2026.10.04

CLXXII

[Renting Out Your 'Next'; Speak in Abbreviations; Pretentious Poetry; Body Language; Language & Love; Meat Puppets; The Great Separation; Language as Truth]

Thesis: Language is not the only way to communicate, but it’s inherently tied to our perception of & ability to articulate beauty & love &, most importantly for our ability to preserve agency, truth.

[Renting Out Your 'Next']

Writing is thinking. Machines let us take action without thinking. We are still responsible for those actions.

Our decisions & actions, and the decisions & actions of others, may results in either our machines or their machines taking advantage of other men & women.

This has always been a concern, well before the advent of LLMs.

Social media feeds are nothing more than machines to which you relegate your ability to decide what comes 'next'.

This is such a bigger deal than one might think - if next is perpetually decided for you... well, you give up your future, one reel at a time.

I know, I know, I'm being dramatic & apocalyptic; I don't think that going on Facebook is selling yourself as a slave to a machine god algorithm owned by man.

But, it's not a stretch to say that such products are designed so that you're perpetually renting your attention, yourself, to such an algorithm.

I digress - this post isn’t a rant about social media. It’s about quite a lot & a very little at the same time.

Thinking, writing, our use of language to express the inexpressible; there's a bit about rap in here, and some about love, too.

[Speak in Abbreviations]

The Death Grips excerpts might seem a bit trite & extreme; perhaps it becomes prophetic when you realize it was rapped on a track from 2011. I wasn't very sentient then, so I don't know, but it seems like the virus he speaks of wasn't quite as obvious at that time.

I myself have been known to speak in quite unintelligible sentences that are a mix of all sorts of slang.

Not even gippity could decipher what the hell I meant

Why do I sometimes communicate like this? I don't know. You definitely feel like you're on the 'inside' when you have a sub language that only certain people can understand.

I remember how cool I felt in my finance days when I would be having conversations about complex options trading strategies, throwing around jargon like confetti when you blow open a pinata. In retrospect, I’m sure some it was annoying and pretentious.

In some regards, though, it is necessary: saying something like “shorting vol at the money & longing it in the tails” is a useful sentence that actually contains a TON of meaning and belief that would be hard to articulate as succinctly without drowning in jargon.

An observation about LLMs is that if you use slang, they'll immediately copy your jargon and throw it back at you. It's almost like a validation that you're on this one man in group. They also write about famous people as if they know them personally, jumping to first name basis and assuming intentions and personality traits with very limited information.

It makes you feel like you're on the inside.

Some of the less professional jargon can feel like linguistic degeneration, but I'm sure you can also argue it's 'just evolution'.

The thing about evolution, is it's not inherently driven towards something 'better.' A robust view of it is as an optimization function pointed towards survival.

What does it mean for a language to survive, though? This isn't obviously coupled to the survival of our current society, at least not on a short time scale. It's coupled to the survival of the language itself, which might be more related to expanding or contracting it’s complexity in relation to the specie’s capacity for linguistic complexity.

[Pretentious Poetry]

MC Ride from death grips is so interesting to me because the running joke on the internet is that him rapping is like "giving a homeless guy a mic.” But if you look at the lyrics, it feels more to me like you gave Dylan Thomas a mic and he shouted everything he had to say to incredible bass.

Dylan Thomas is one of my favorite poets. He does this cool thing where he mixes words together in a way without a literal definition, but have an intuitive meaning that shines through:

What does it mean for blind eyes to blaze like meteors? Nothing literal I don't think, but you get what he's going for and it's beautiful.

Some of MC Ride's lines are like this, too. There are lots of Biblical, or even Cosmic scale allusions, spoken in now uncommon sentence structures.

This is so interesting to me because if you listen to a Death Grips song without processing the lyrics, it sounds kind of like disoriented, angry shouting. Which, on it's own, is great music for shipping code, but it's somehow better given the nature of the underlying lyrics.

Both him and Thomas show how you can use language more than literally to convey meaning.

[Body Language]

I'm sitting at a different coffee shop to write this than usual.

If you've ever met me in person, you'd be quick to observe than I sit in all sorts of strange positions, oftentimes sprawling out. On booths, a lot of the time, I'll put myself in criss cross apple sauce.

I was doing that at this new coffee shop, and a guy who works there was walking toward me with a look in his eyes and a body language that says, "I'm going to tell him he's doing something wrong."

Perhaps it was a mix of where his eyes were looking, the authority he was trying to convey without having too high of urgency, ample experience on my part of being told to take my feet off items before... whatever it was, I accurately guessed he was going to tell me to take my feet off the bench before he actually said it!

That's not to say that I think me guessing this was impressive, it's really not; the fact that it's not impressive is why it's interesting.

Some communications don't need language.

[Language & Love]

There are some human emotions, like love, that are impossible to convey exclusively with words.

I think this makes sense. Our species has likely felt some version of love for longer than we've had the language function. I'm sure there's an argument that love is 'more' or better now that we have a more complex language to express it & support it, but the underling biological machinery is still there.

There are some things, like a hug, that I don't think language can capture in quite the right way. Or actions in general, like acts of service, or gifts.

You can say, I want to hug you, and this can make someone you care about feel quite good. But I think maybe this reaction is a sort of anticipation or imagining of the hug - the hug itself is obviously much better!

Sacrifices, too, transcend words. Giving resources, time, attention, or doing things for someone that you wouldn't do on your own are all things you can say you’ll do with language, but actually doing them makes the difference. Those words have a stronger impact, too, when there's a track record of actions that convey the emotions the words are trying to capture.

That said, words can still be expressive forms of love themselves, especially when they imply that you’re genuinely thinking about the person.

And then, even beyond that, there's always that deep feeling that you wish you could give or express more than you already have, that question in the back of your mind, "Does she really know how much I love her? What could I do to express it more?"

So maybe, when it comes to something like love, it's not that language feels insufficient. Rather, it might be that everything feels insufficient.

[Meat Puppets]

There's a term going around that I shockingly heard someone wear with pride: He called himself a 'meat puppet'.

This means that he views himself as an intermediary for an LLM. When I send him text, he just sends it to his LLM. When it asks him a question, he sends it to me.

So, what'd you think he decided to do? Well, he just let me talk directly to his LLM instead of him!

This is not SciFi anymore, people are actually doing this.

I think it's weird. I guess, in a way it works? Because now, he doesn't have to do anything except approve decisions?

There's an objective truth here: it's more convenient for me to talk to his LLM than it is for me to message with him if he's just going to dump it into his LLM anyways.

Part of me is shaking my fist at the sky, but at the same time, I can see the other pressure: if it more efficiently gets us both what we want, and we control for the risk... is there a problem?

I could ask who controls the 'thinking machine' he's learning to depend on; but so to, you could ask who controls the 'thinking machine' I write this on? I own it, and the os is open source... but I couldn't make myself another cpu if I needed one...

While it all reeks of a Faustian Bargain, perhaps we've been making such deals since first man traded his fist in for a bludgeon.

[The Great Separation]

I used AI a lot more than usual to code faster this weekend, but I was still 'driving.' It was a quite sprawling project that touched every layer of my app; I would not have been able to cover nearly as much ground as I did without AI.

I still view it as a tool, though. And I view it cautiously, and almost with a fear that it presents a trap, an easy way out of everything that could decay my own capabilities if not used carefully.

People say that there will be a separation: some people will use AI to amplify their output; others will become increasingly dependent and less intelligent because of the tools.

Don't we all think we're going to be the ones to amplify our output? Whose right?

If the code makes me more money, am I defacto right? Would seem to be evidence, not proof.

Is it maybe about treating it, treating all technology, as tool that amplifies your output without compromising your judgement?

But what does that actually mean?

I'd say it should be intuitive, but I've seen far too many people fall into the trap of AI Psychosis: the models output words that convince them they are right and great and the best and correct, when these phrases are meaningless or castles made of sand built on entirely false priors.

I've explained the story of how an LLM & I built a narrative together of this 9 figure company being steered by a PE investor. Then, when I brought up the investor to the CMO, he said something like, "Whose that? Oh, them. Yeah we don't talk about them ever."

I was convinced I was going into that call with a valuable POV. It was bullshit.

Maybe it is as simple as saying you need to preserve your ability to think for yourself. Identify & doubt priors, cross reference everything, ignore empty words not backed by reality. I’m still cautious, and think you should be too.

I’m hear every week about founding, epistemology, ai, writing, and the rest. Subscribe if you want to hear more.

[Language as Truth]

Maybe it's this: Language can be used to convey truth. That truth is testable against reality.

When we use words, communicate ideas, we have a responsibility to make sure they're lined up with reality. Even, every once in a while, nonsense words can articulate that for us.

We have to be supremely careful that we don't sacrifice our critical faculties to the machines. If you do, you really sacrifice it to the men who control the machines.

If you don't have a mastery over language, and aren’t on guard to check claims against reality, I think you can be more easily fed lies that sound true but are detached from reality.

Compression of language, abbreviations, these aren't inherently bad; I’m sure they're worrisome when they're an artifact of being forced into a stream of 'next' that you no longer control.

And, what of the musings on love & poetry?

Well, language lets you wrestle with and attempt to articulate the intangible, even if it doesn't always capture what you hope for.

There's beauty in that, too.

Live Deeply,

2026.09.27

CLXXI

[Take the Flight; And Then There Were None; Changing A Big Decision; Self Documenting; The Snowflake Effect; Tail Bets; No Permanent Plateaus]

Thesis: Being patient now so you don’t have to be patient later

[Take the Flight]

I took the most spontaneous flight I ever have this weekend - booked around 00:30am Friday, taken around 01:00pm that same day.

I don't regret it at all - in fact, it may be one of the best decisions I've ever made.

I still don't have fuck you money, but I'm also fortunate that both my cofounder & I agree that there's no sense in waiting til we're 'done' to occasionally exercise the freedom we're building for ourselves.

Still, I have been thinking about the notion of being patient now to be impatient later.

A few months ago, I said this was happening in my business. Now, it's accelerating.

This post will still be about B2B SaaS today, in a bit of a more non sequential way than usual. Don’t expect a grand revelation or through line; rather, it’s many meditations

But, first, I must say that even outside of business, I was being quite patient in a respect that I didn't even recognize as patience, because I didn't know what I was waiting for anything.

I don't know anything about you, dear reader, so I’m uncertain if you need to hear this, or if you should, but… when the timing is right, and you know it, you should take the flight without hesitation. You'll be glad you did.

[And Then There Were None]

Two more of WhiteWhale's competitors disappeared in the last 2 weeks.

The first one was interesting, because it was a company run by a non technical person who was deeply AI pilled vibe coding software, who all of a sudden just threw in the towel. I know the guy, and I know that he's pretty good at sales and distribution.

I think his venture lasted 3 months - an interesting counter point against the narrative of "You can vibe code anything and sell it as long as you know how to sell." I still think that it is possible to vibe code and sell a SaaS, I know people doing it, but I think the notion that building useful products and successful businesses is only dependent on 'distribution' misses a lot of nuance.

The second competitor had their team was hired by HubSpot. I'm not calling it an acquihire, as the press releases strangely did not call it an acquisiion, although some money likely changed hands, because an early investor is claiming they got ROI on the deal.

HubSpot has been very acquisitive lately, and is very aggressively building new products & features out. I think in another 2 years, it will be very interesting to see how much this translates into retained and expanded revenue.

I don't really know what's going to happen in the space, and I don't even feel comfortable making much of a strong prediction. That said, perhaps I'm naive or arrogant or both, but based on what I've seen so far, I'm not shaking in my boots about big companies all of a sudden being able to make way better software 'because AI', nor do I think that acquisitions & the like become less value destructive 'because AI.'

I think the structural advantages of a startup still exist, and if anything are more amplified by AI than the advantages of large companies are.

Eg, the binding constraint on a big company's code base is NOT the speed code can get written, it's the speed at which PR's can be merged, systems integrated, and committees convened. Maybe if you just yeet all of that to AI the speed does increase, but so does bloat.

Then, in terms of competitors my size... well, it's really hard to make a profitable, sustainable business in our space, and we are seriously starting to wonder if we are the only pure play 'signal company' that is profitable.

All of this is to say that it renews our vigor and energy to push the advantage while we have it.

It's the sort of thing that helps you stay in the mindset of "How do I achieve this goal for the customers as quickly & best as possible without breaking anything" rather than "I don't know if we should do this what if we break something or they don't like it."

You start to trust in your ability to predict that more and more, while accepting you can still be wrong.

[Changing A Big Decision]

We're writing a feature for WhiteWhale that we've explicitly said we'd never ever do. Crazy, right?

I still think it was right for us to not do it when we didn't do it, and I think it's right for us to do it, or at least try it, now.

I have a pretty mid version of it done, but it might take a week to make a really impressive version that gets us a lot closer to not only the thing our customers say they want, but the thing we see them doing even without our product (the latter is so much more important than the former).

It's a big risk, and it's basically impossible to get it 'perfect', but wanting to get it perfect is one of my toxic traits that slows down great.

This feature will be an attempt at completing the next 2 or 3 steps in our client's workflow. Even if it only succeeds 50% of the time for them, that is actually an insane improvement over what we have now, and very likely good enough that we'll see an uptick in usage, and more importantly, ROI.

[Self Documenting]

You would be surprised at how many things are not as self explanatory as you'd think. Even if you really think you're spelling something out for users, it's likely that it's not clear enough.

Take the below water faucet. Undoubtedly, entire committees argued about what icons they should use for the 4 buttons on the thing.

Inevitably, it still confused a high enough percentage of the users that Delta felt the need to print out a little explainer sheet, defining what each icon meant.

I remember the first time I used one of these and thought the sparkling water button was “poison gas” too

God, wouldn't it be so much easier if there was a design that even 99.99% of people just intuitively got?

During my time writing software with only hundreds of users (I suppose the total number is over a thousand now), I am shocked again and again by how much simplification is needed.

Changing copy here, removing a button there, has a real impact on getting people to take action. As an example, we just overhauled the copy & formatting of the Slack alerts we sent out and we're already having more users tell us about wins they're getting by following the alert to take action.

[The Snowflake Effect]

I once heard an argument that half the reason CTO's buy Snowflake is because they get to be the smart CTO who bought Snowflake and tell all the other CTO's about it at conferences. It becomes part of their identity in an additive way.

We want WhiteWhale to have this effect. One of our partners actually just posted about us on LinkedIn in a very natural way that makes me think in that direction. You can check it out, but basically it is "Signals are really important for outbound. You can either build your own tools or buy WhiteWhale, there isn't a comparable option on the market."

This is great for both the Partner and us: We get a partner bragging about using us, and he gets to be associated with the sexy new cool up and coming thing.

It translates to revenue for both of us, because it helps him get customers, and when he does a customer, there's a decent chance he grabs another WhiteWhale license from us. And better yet, because we know he's really good at using the software, we know that we won't have to do much Customer Support and the client is more likely to stay for longer.

This isn't our first partner like this, and we really don't want it to be our last.

[Tail Bets]

We took an interesting assymetric bet many months ago with another company that wanted to resell our data in their platform. It required quite a bit of engineering on our end, and I doubted if it would work.

For a while, it didn't. But, we changed the pricing model with them to be entirely consumption based. There had been so little usage that we weren't even charging them for it, but this month, all of a sudden, usage ticked up to what I think will be 4% of our revenue for September. This is really cool, and is a nice win.

I think I'm sharing this because it's a cool win and there are a lot of variables that go into getting something right. Changing something as simple as pricing can have a massive impact.

And, at this rate, anything can happen, but I wouldn't be surprised if it became considerably more than 4% of our revenue over time and out stripped our other growth.

If you enjoyed this post, I’m here each week talking about bootstrapping SaaS and living life and, when I’m really in a mood, preventing the fall of the west.

[No Permanent Plateaus]

I love the idea of always having 'another level' to break through.

At first, when you have no customers, or get stuck for a bit, you might lose hope and feel that these levels are insurmountable.

It's counter intuitive, but I really do think it's important for you to try and 'master' the level before you spam through to the next one.

I talked about plateaus briefly before; here, I reaffirm that I think explosive growth happens because it's very hard to know what the real binding constraint is, but it's not so hard to know what the 10 most likely binding constraints are.

So, as you aggressively solve all of them one by one, everytime you're wrong about the problem you solved being THE binding constraint, but are right about it being one of the next 10, you're simply clearing the way for a growth spurt.

Because, once you solve the real constraint, if you've already accidentally solved the next 5 in an attempt to solve the main one, well, now you have a clear path to grow grow grow like gang busters.

I think there’s a similar pattern with a lot of things in life, which is maybe why it sometimes feels like so much good can happen all at once.

Live Deeply,

Noah Jacobs

(Signature image absent today as finishing post on cellphone, will be back)

2026.09.13

CLXIX

[Binding Constraints; Running a Coffee Shop; Limiting Beliefs; Pain is Non Binding; Google Search Sucks; When it Clicks]

Thesis: Growth is about repeating the process of identifying and removing the binding constraint; when you do, all of the other hard work you’ve done pays off at once.

[Binding Constraints]

Building anything awesome is a series of making progress, hitting plateaus, and then figuring out how to get past the plateau.

When you want to achieve some goal, often times, there is usually ONE thing that will move the needle much more quickly than ever.

There are other things that you can do that are good and important, but, if you’re actually caught on some other constraint, you won’t see the gains until after you tackle the real constraint.

This doesn’t mean you should do analysis paralysis and not work on anything because you don’t know what the binding constraint is, but it does mean that you should be vigilant about trying to find it.

Because, when you do, it feels like something all of a sudden clicks, and you grow in a rapid step function rather than incrementally.

It’s important that you don’t lose hope at the plateau, because assuming you’ve been working on objectively valuable things, the longer you’re at the plateau, the more you will rip once you solve the real problem.

This is important to remember, because finding the real constraint is super hard! But I do think that the longer it takes, the more worth it is - it will let you realize the gains from all of your other good but non binding changes all at once.

[Running a Coffee Shop]

Say you run a coffee shop. You have one barista and one machine and one cashier and altogether you can serve 30 people an hour.

You’re stuck, though! You can’t increase revenue! You keep marketing and marketing and trying to get more people to come, and there’s so much demand, but you’re stuck! You feel like you’re running in place.

Then it hits you - you’re at capacity! You are serving 30 people an hour and turning away another 15.

Marketing wasn’t the problem; your capacity was!

So, you double capacity. Now, you can serve all 45 people, and your revenue 1.5x’s immediately.

It feels like something clicked. The one single change, increasing capacity, let you unlock the gains from your previous marketing efforts. In this way, the gains from your previous hard work was finally realized all at once.

It doesn’t mean your marketing efforts were for nothing, you just couldn’t realize the gains from them until you tackled the actual constraint.

[Limiting Beliefs]

A lot of the time, limiting beliefs are a binding constraint.

Meaning there's one thing that if you did, you'd have a great outcome, but you keep not doing it, because you don't actually believe or realize that it's possible for you to do it.

This weekend, as an example, there was a specific data source we wanted to scrape for WhiteWhale, but were avoiding. The source is notoriously challenging, and back in maybe January 2025, I made an attempt on it and failed.

So, I had a 'prior' that this site was not worth it to scrape.

Last week, we were evaluating vendors to do it for us, but Jack actually challenged my prior and asked how hard can it really be. He was right to ask that, because I took another stab at it and did it in maybe 4 or 5 hours of coding time yesterday - not so hard at all.

This put the cost at ~1/5th of what the vendors were quoting us at, and we can get it even lower, really. There will probably be a little maintenance and editing to do over time, but at the volume we wanted to hit, this is worth it.

So, in the context of acquiring this data, my limiting belief was actually the binding constraint.

This is hard to realize sometimes, because it is often times efficient to come to a conclusion and then stop thinking about it for a while as you move on to the next thing. But revisiting your priors can be super powerful to see where you might be wrong. Especially when 1.5 years elapse and you’re much more competent in some regard than before.

[Pain is Non Binding]

I read this beautiful book a week or two ago, "A Canticle for Saint Leibowitz". It follows a monastery in Utah as the world recovers over 1500 years after a nuclear holocaust. In the last stage of the book, when society is more developed than it was before the disaster, the Abbot of the monastery reflects on pain after talking with a Doctor who wants to euthanize patients in great amounts of suffering.

The implication of suffering & pain being ‘evil’, is that absence of suffering and pain, and possibly even pleasure and comfort, becomes good.

This is directly antithetical to what I believe about achieving anything worth while: inherently, valuable things require pain and discomfort to achieve. Especially when you’ve been working towards something for years and hit a plateau and are going crazy.

If you believed discomfort was bad or evil, you’d probably be inclined to give up! That’s really easy when pleasure and comfort are so cheap.

Now, this might offend your sensibilities, but good and evil is not a spectrum where suffering is evil and pleasure or comfort is good.

Suffering is not categorically evil, and pleasure is not categorically good.

Let's say I'm sitting down and coding and really in on something and I kind of have to go pee but I'm locked in and want to finish the thing. The feeling of having to go pee is causing me slight discomfort, so okay, I'll go pee. Okay, now I sit back down and you know what I don't love the way the screen is hitting my eyes in this lighting, let me change the color scheme to something else. Okay great, but also, I wanted to listen to this one song while I was coding, let me pull up YouTube. Okay but yeah watch this quick recommended video first. Okay and now I’m at the song but this song was not like really the one I wanted to listen to let me swap it. Oh okay yeah but my water is out let me go fill it up. Okay great we're ready to go but you know what I'll probably do better if I have some sugar let me grab some honey. Okay great but now a weird song is on let me change it.

Do you see the problem here?

Doing the work is never going to be comfortable! Your comfort (or lack thereof) is not the binding constraint.

Someone once told me that he would be able to do SO MUCH more work when he bought an ergonomic keyboard and a better chair.

I'm sure it's nice, take care of yourself, but it's not the fucking problem. If you're not working "because" your chair is ‘sub optimal’, you're actually not working because you're avoiding the discomfort of starting something hard.

Optimizing everything becomes a disease the minute it distracts you from solving the binding constraint.

Imagine saying “I shouldn’t have climbed this mountain because it hard and uncomfortable.” What?! Are you crazy?! I am so fortunate to be in a position to be able to overcome the challenge - so why not!?

[Google Search Sucks]

Google search sucks, and summarizing the results with Gemini doesn’t fix it.

Search sucks because it’s easy for people to manipulate google’s search engine to get the content they want near the top of the results. Because you can make money by having your results near the top of search, it is worth money to have them there.

So, people who can make money by having their results there will spend money and resources to get their results there.

This makes it harder to use Google to find the information you’re looking for, if not damn near impossible.

One example I dealt with in mid 2025 was integrating WhiteWhale into our clients HubSpot & Salesforce CRMs. There are a lot of company that will charge you money (think, $2K or more a month) to do this for you. So, they spend resources flooding the search engine results with articles that explain how hard it is to do this integration yourself and how their solution is the best.

In a classic example of Goodhart’s Law, because Google has a formula for estimating the most useful results, the inputs to that formula stop being useful ways to estimate the most useful results - they’re gamified instead.

This is an underlying problem with search that needs to be fixed. By focusing on summarizing the results their search engine already produces with LLMs, I think google is working on the wrong constraint. Summaries of junk are still junk.

I think this is probably why a lot of people, myself included, use things like GPT for search that go further than summarizing and do multiple searches and follow links. But even these have problems that will likely be exploited more and more over time.

If you enjoyed my reflections on growth, I’m here every week, sharing my thoughts and what’s working for me as I grow my business and try to become a better person.

[When it Clicks]

All that is to say, you shouldn't get into such an analysis paralysis because it's not obvious what the most important thing to change is.

As long as you can find something that will obviously move the needle, even if you think forward progress will be slow, forward progress of any sort is preferable to getting stuck.

We had a wrestling coach visit our jiu jitsu gym the other night. My wrestling defense is okay against jiu jitsu people, but bad against wrestlers; even if I can sync a guillotine, if the person is decent at wrestling and jiu jitsu, they still might pass.

I spent 45 minutes with this coach, though, and my wrestling massively improved all at once. I used to like kind of half ass grab a single leg and then not know what to do with it. I'd over think some specific sweep, but he gave me the heuristic that transcends all: keep pushing into the opponent. Then they kind of just fall!

All of a sudden it clicked, and I swept a guy I'd swept maybe once in half a dozen matches 3 times in a row!

That said, I’ve probably been told what this coach told me in the past, but it just didn’t click. That probably just means that now I happened to have the pieces and training already in place to rapidly internalize that advice. So, even though I had missed this binding constraint for a while, it surely wasn't always the binding constraint, and the other wrestling work I did primed me to have a step function increase.

In this way, it was not all for nothing.

This has happened to our business, WhiteWhale, a number of times: intense product & positioning work followed by one change that failed in the past, like increasing price or allowing for self service, that now all of a sudden works because the other pieces that used to be missing are now in place.

As aggressively as you want to look for the binding constraint, even if you don’t have it, keep doing things that you know will help. Then, when you do find the constraint, you can grow really fast all at once. This is what it means when something just “clicks.”

Live Deeply,

2026.09.06

CLXVIII

[Being Great; Escaping From First Principles; Perpetual Beginner; 2 Years of Failing Up; Low Trust; Anti Bullshit; Beautiful Code; Spaghetti Code; The Power of Seeing Different Levels; Don't Lose the Art]

Thesis: Perhaps mastery comes from seeing what is possible and being able to derive it from first principles more than it does from pattern recognition.

[Being Great]

I want to be great at the things I do.

That is really hard to do. It's easier to NOT be great. It takes a lot of concentrated and consistent effort to be great.

I saw my jiu jitsu instructor, who has been a black belt for over two decades, discover the defense for an uncommon move from first principles in real time the other day.

I think that this is somehow related to greatness, perhaps, or mastery; being able to approximate truth from first principles on the fly.

Getting to that point is super hard, though. Constantly feeling like a beginner must be an important ingredient for it, and building the confidence to know & be able to re derive things from and handful of basic truths.

My post last week, I didn’t really dig into what I think great or beautiful code is. I’ll try again this week. Unfortunately, I don’t have a formula, and can’t prove it, so I’ll telling you why I think it’s self evident that it’s beautiful, not how I can prove it.

Finally, an anecdote from C House to drive home why seeing what is possible is so important to being able to do it.

This post is a bit winding and reflective; I’m trying to articulate these ideas for myself as much as I’m trying to share them with you.

[Escaping From First Principles]

In the Jiu Jitsu leg lock lab I went to on Friday, I got to watch a 20+ black belt teach an absolutely lethal submission I was completely unaware of. And then, when I asked him how to get out, I got to watch him say, "I don't know," followed by 20 minutes of real time experiments before he formalized a succinct, useful answer.

First off, the submission was a short ankle lock with a butterfly hook in. I have been grappling for nearly 5 years and well over 1000 hours, and I had not once in my life seen this exact move before. This is not an uncommon experience in BJJ, to learn something new so far into your journey: it constantly reminds you that, in a lot of meaningful ways, you are perpetually a beginner.

This submission is such an interesting one, because it contradicts a common truism: everyone teaches that if you are in a leg entanglement, and you 'clear the knee line', you are safe from further leg submissions for the moment. I was even taught this implicitly at John Danaher's gym, who is one of the guys who revolutionized leg locks for the sport.

This short ankle lock makes it simply not the case: when an opponent is escaping from ashi garami, you butterfly hook the inside of their leg close to their knee with the leg that was your outside leg, then cross over your other leg to capture their knee.

If you don’t do jiu jitsu (even if you do) that sentence might not make any sense. If doesn’t matter, because they've 'cleared the knee line' but are in imminent danger of a wicked submission.

I asked our coach how to get out of this. More crazy than me not having seen this submission is that this 20+ year, 5th degree black belt didn’t have the escape memorized. He did not lie and say that he did - rather, he said “I don’t know,” then proceeded to spend 20 minutes figuring it out in front of us.

He had an experienced player repeatedly sub him with the lock until he 'discovered' that you need to grab their non attacking arm and pull the elbow toward you without yanking the hand up and out, as that would let them finish the sub faster.

I don’t know if I can overstate how sublime this was to watch: the man was using his deep understanding of the game to arrive at an accurate answer in real time.

He didn’t need to watch a youtube video or go to reddit; he understood the principles so deeply that he was able to independently arrive at a correct answer.

[Perpetual Beginner]

Lately, I have been feeling like a perpetual beginner in everything I do.

When you start a company, you always have some challenge. And when you solve that challenge, if you want to keep growing,you have another challenge immediately after to deal with!

And it's very likely that it's a new challenge or a new non obvious version of an old challenge. Otherwise, it wouldn't be such a challenge, or you would've solved it pretty quickly.

And engineering wise, man, I always expect more of myself. I have an impossibly high aspirational standard that I'm never going to hit. It's easy to forget, though, that you've been getting closer to it, especially when you have your head down for years at a time.

You might have a moment of hubris that gets flattened by some risk or hazard previously unknown to you. The hubris stays, in the sense that you gain the confidence that you will figure it out. But, it’s less that you already are so great and know the thing, and more of an admission that you have no fucking idea what the answer is but WILL find it.

When I was in Greece on a ferry, I was thinking, "What would I do if this boat sank and I knew no one was coming to rescue?" I could always see some coast, even if it was miles away, so I thought, "Oh, I'd just swim to shore."

And I genuinely believe that I would be able to if I needed to! Maybe in reality I'd die, but I'd die trying. But you don't think about that part, and you wouldn't if the situation arose, other than maybe to say, "If I died, what would be the most likely reasons and how do I stop that?" Then do anything in your control to prevent it, and keep going regardless.

Something like starting a business or learning a new skill is like that for me. A weird walking contradiction of "I have no idea what I'm doing but I'm going to do it."

[2 Years of Failing Up]

If you spend so long thinking like this and look up for the first time in a while, there is a certain feeling of shock at how far you’ve come even though you still feel like an ignoramus.

Last week, I found myself explaining to someone the difference between server side and client side rendering for websites and showing her how to watch network calls to learn about a site and how it’s designed. Afterwards, I realized how much more I understood about the internet than I did, say, 2 years ago.

Likewise, NL explained to me his dream product for some analysis he wanted to do with a bunch of job postings he had, and I gave him a very straightforward solution involving a relational database that is quite similar to a pattern I use to process hundreds of gigabytes of data regularly & efficiently. This wouldn't have been so obvious to me years ago either, but now it's instant.

That said, the line between pattern recognition and first principles is thin in the latter example; perhaps it doesn’t matter in the average case, and only matters in specific ones.

[Low Trust]

I'm an incredibly low trust person. When I was in Greece, it took me 5 days to realize that a lot of the people, including the owner of the hotel I was staying at, were just genuinely friendly and not trying to secretly take advantage of me.

I write this blog in the same cafe in a quite safe area of Cambridge every Sunday; when I go to the bathroom, I take my laptop and phone with me without fail, because I don't trust anyone enough to watch them for me.

When someone tells me that something is true, I often times push on it and see how deep their belief is in the thing.

When they get stumped, the most trustworthy people will say "I don't know," and then try to work it out in real time, like my coach did. This happened recently, both when I was talking to C House residents about genetics & quantum: you can often tell when someone is actually working from an internalized understanding of something rather than simply pattern matching to things they know.

I don't feel shame very often anymore, but when I do, it is almost always when I catch myself saying something confidently with insufficient information. I really don't like bullshitting. Sometimes, you say something that you think is true, but as you say it out loud, you start to question it and realize it's maybe not so true. This also happens when you write, which is one reason I like writing so much.

This is why the whole thinking from First Principles thing is so important: there is some set of very basic things that you assume are true*, and then everything else should be derivable from them.

I don’t think I’d have to tell you that this cat has beautiful eyes? Somehow you might just know it.

If you operate from First Principles, you can operate effectively from a low trust position. The other person should either be able to recreate their claim from very basic things, or point to repeatable observations, rather than just saying something like "Oh I just know it's true" or "So and so said so!"

Perhaps thinking that the hotel host has knife behind his back is too far, or perhaps it's still the best strategy; guilty til proven innocent, as so many things are.

_{*Of course, these basic things can be wrong, too! Such as our understanding of time as a constant was before the theory of relativity. That said, you try to keep them as basic as possible and revisit them if contradicting information comes up, which is easier said than done.}

[Anti Bullshit]

I think one of the reasons I get so frustrated by LLMs is because they can't obviously hold a model of cause and effect or deterministic rules in the same way we can. And, of course, they bullshit a lot, too!

When a human bullshits you, you can call them out. Calling an LLM out is something that isn't so effective, other than to maybe adjust prompting or a harness, which is still non deterministic. Sure, you can do things like guarantee structured outputs, or could make a complex system that checks citations automatically and flags broken links or links to sources that are semantically dissimilar from the point the LLM is making, but this still isn't deterministic in the sense of this is consistent within a specific worldview.

Perhaps I need to be more detached, and find a way to use them for what they are. But since I know this about their nature, it’s quite frustrating when I depend on one and it gets something wrong for this known reason!

When it tells me something unverified and I believe it and I make a decision based on it I feel like an absolute fool and utter mark.

"Why did I trust that thing to make this decision?! I should've checked!"

Yeah, you should've checked! But when it's so easy not to, it is even harder to check. It's quite easy to get in that state that isn’t flow where you're not really focused when you are doing deepwork, but you feel productive and are paying enough attention but not all of the attention. You let yourself get distracted and those moments of distraction 'feel' productive but aren't.

In all honesty, this is one of the reasons that I don't keep an LLM directly in my IDE or 'terminal': I want it to be harder for me to depend on it and to let it drive.

I actually think that friction is good, because I even see when I am coding with it open in another window, there is a pressure to just use it and nothing else.

[Beautiful Code]

NOTE - this section is more technical than I usually try to write. Regardless, I’ve made an effort to make it clear.

In last week's blog, I think I failed to really drive home the difference between garbage code, okay code, and beautiful code. If I'm unable to show you that, then I don't think my other arguments make as much sense.

Luckily, I do think that when you reflect on it, the beauty of certain code becomes self evident! I wish I could prove the beauty, and maybe one day I'll be able to. For now, I'll give you symptoms of it.

Beautiful Code is easy to maintain, edit, and extend:

• Maintain: Things don't break so often, but when they do, you can quickly identify and isolate what went wrong and fix it.

• Edit: if you decide to change an assumption or edit some behavior, you can quickly pinpoint want needs to change and change it without breaking a bunch of things

• Extend: If you want to add new functionality or iterate on the code, a lot of the core plumbing and tooling should stay intact; adding one more function shouldn't break all of the others!

Now, below is some code I wrote for WhiteWhale that I believe is beautiful. Basically, WhiteWhale sends out a lot of notifications (too many?) to users via Webhook, Slack, Microsoft Teams, and email (mainly email). This pipeline is what we use to identify users who are supposed to receive these emails, grab the data for them, render the email, and send it to them.

    async def email_update_pipeline(self, filter_users: List[str] = [],
        mode: Literal["Base", "Admin-Assignment", "Admin-Overview", ... "Search-Results"] = "Base", tz: str = 'et'):
        """
        ....
        """
        if mode == "Base":
            functions = [
                (self._get_target_users, []),
                (self._add_top_articles_to_users, []),
                (self._add_opportunities, []),
                (self._write_html_template_email, ["signals.html"]),
                # (self._write_emails, []),
                (self._send_emails, []),
                (self._mark_articles_as_sent, [])
             ]
        elif mode == "Admin-Assignment":
            functions = [
                (self._get_users_with_specific_active_emails, ["admin_assignment"]),
                (self._get_*, []),
                (self._get_*, []),
                (self._eliminate_*, []),
                (self._write_html_template_email, ["admin_farsight.html"]),
                (self._send_hermes_emails, ["*"])
            ]
        elif mode == "Admin-Overview":
            functions = [
                (self._get_users_with_specific_active_emails, ["admin_overview"]),
                (self._filter_users_on_*, []),
                (self._get_*, []),
                (self._get_*, []),
                (self._get_*, [False]),
                (self._write_html_template_email, ["admin_overview.html"]),
                (self._send_hermes_emails, ["*"]),
                (self._mark_admin_overview_as_sent, [])
            ]
        .... more functions ....
        elif mode == "Search-Results":
            functions = [
                (self._get_unsent_searches, []),
                (self._get_search_data, []), # filter out any w no data here
                (self._write_html_template_email, ["search_results.html"]),
                (self._send_hermes_emails, ["*"]),
                (self._mark_searches_as_sent, [])
            ]
        else:
            logger.error(f"Invalid mode for email update pipeline, {mode}")
            return

        current_data = None
        logger.info(f"Beginning Email Pipe with {mode} mode enabled")
        for function, base_args in functions:
            logger.info(f"Email update pipeline processing {function.__name__}")
            if current_data is not None and len(base_args):
                result = await function(current_data, *base_args)
            elif current_data is not None:
                result = await function(current_data)
            else:
                if len(base_args):
                    result = await function(filter_users, *base_args)
                else:
                    result = await function(filter_users, tz)
            if result.success:
                current_data = result.data
                logger.info(f"Email update pipeline completed {result.function_name} with {result.result_count} results")
                # print(result.data)
            else:
                logger.error(f"Email update pipeline failed {result.function_name} with {result.errors} errors")
        return current_data

Used * to replace some names. … marks missing docstring plus more function lists. Rest is verbatim.

Some things I'll point out that help with making it beautiful:

• Data Driven: The actual logic is ~ 20 lines at the bottom; in effect, the if / else branches are the same as a dict, and I can and should replace it with such. I really thing it is nice when the data structure can drive the execution of the code

• Low Dependencies: this part is very much plumbing, so we shouldn't really need many outside dependencies. Other than my own functions, I think logging is the only dependency? There are more below, but if I need to nuke one, I don't have to rewrite this core part of the code.

• High Level of Abstraction: This is hardly even code! It's very clearly a list of functions. The only 'coding' I'm doing here is deciding a valid combination of functions! The 20 lines below that have been executed 4 or 5 times a day for over a year without problem!

• Consistent Structured Output: Every single one of these functions returns the same data structure. So, we can succinctly get a consistent 'status update' at each step of the pipeline. This amortizes logging cost.

• Error Handling: The structure is an envelope that catches most failures and brings them up to this level of code; rather than getting a raise that could break some other shit that I don't want to break, the function just tells me when it broke and moves on (this works well here because very rarely if ever is the other code being executed path dependent on this code)

There are problems with this code! But, when errors happen, I can almost always pinpoint the exact failure point rapidly.

And really, the whole thing really doesn't need to be in a class - it sort of feels convenient for holding some constants or configs, but I could do just as fine with a dictionary of such things optionally updated at runtime with arguments. Probably would be better, because I wouldn't have to worry about storing state in both the instantiation of the class and the function call!

Also note that I didn't START with the code like this; the first sending pipeline, if I recall correctly, was standalone. It wasn't until 2 or 3 that I added the lists of functions. And then it wasn't until more than this that I generalized the template filling function and the user selection function. And in some cases, I didn’t bother to change the old functions to use the new general ones, which is lazy and/or efficient.

It was very gradual and incremental and updated maybe not when it was 'obvious' that it should be update, but perhaps just before it was 'obvious.’

Now, if I want to add a new pipeline of this type, it is actually really fast. I perhaps need to write one or two functions to fetch and mutate data, along with the list of functions. And, an email template, which I’ll have an LLM write once I decide on the underlying data structure to feed.

[Spaghetti Code]

Writing code like I showed above is not only joyful to me, but it also creates massive speed gains in terms of my ability to write it.

This is because I've basically raised the level of abstraction on which I'm operating for 80% of the 'work' in this function. Again, adding a new pipeline is typing out a list of functions, maybe two of which I have to define.

I don't have to write a new sender or new html template rendered or a new function to figure out what user's to get, because I've progressively generalized this.

Now, the worst form of that code might be if I rewrote every function for every instance of the pipeline I needed. If I did that, there would be a LOT of code. Something like 2 or 3 times as much! This would be a lot to maintain and pretty bad. If I fixed the sender in one spot, it would still be broken elsewhere.

A more common form you might see would be to have the most obvious things, like the email sending and user selection as helper functions. In my opinion, this is WAY better than the awful form I just mentioned, but not as great as the above form I showed, because you'd need to write a standalone function for each of the instances of the pipeline, and you'd have a lot of redundant code. You might also be less inclined to be disciplined around the consistency with the input / output data structure. Overall, it would be quite a bit harder to hold the whole thing in your head as it grew overtime.

The code above is designed to be held in your head regardless of how many pipelines you add. There is a lot of powerful consistency.

When your code does cross that gap though to the realm in which you can hold enough of it in your head to fully understand what you're looking at quickly, and quickly debug it, and edit it, and extend it, that is where I start to think that speed gains with an AI system are not so relevant because your writing speed is already asymptotically fast.

Perhaps there is a boost from the "parallel processing" in terms of writing you get to do when you have an LLM running on one task concurrently while you run on another, but for a sub 15 minute task, I wonder if the startup and tear down time of that process, as well as your brain's switching cost, kills the gains?

And, if you say, Noah, but my code base is not there yet, I'd wonder: will it ever get there if you let the LLM do the whole thing? Maybe, maybe not! Again, from my own experience using them, they’re like a Siren call towards a normal, average code base that you lose comprehension of overtime.

Remember, nearly everyone who is a complete, die hard vibe coder says they check the code at first and then stops altogether!

[The Power of Seeing Different Levels]

I think that if you build something cool or that you think is beautiful or even just effective, it's nice if you share it with the world. For one, if you're wrong, you want someone to be able to tell you you're wrong, and why, so you can grow faster.

The other reason is that if you're right, you're paying it forward: It's hard to know what performance looks like until you see it.

One of the new residents of C House, PD, said something that shocked me last night:

This came after myself and NL we're talking with him about sales and go to marketing strategy.

PD sells a hardware product to physical trainers and has done quite impressive numbers very quickly. A couple days ago, he asked me how I would scale up if I were him. I just asked him questions about where the bottle neck was, and it sounded like it meetings booked, because he already has a 30% win rate, which is epic.

He asked how he could get more customers, and I asked him if there was a database of people with licenses he could sell to.

Then, last night, he said, "Man, I found the database, but it's not useful! It only has the trainer's LinkedIns, and people don't respond on LinkedIn in my space!"

NL and I started laughing out loud, because the solution was so glaringly obvious to us. NL: "Oh man, I'd also be so pissed if found a database and it was full of absolute gold!"

PD didn't know this, because he simply hadn't been exposed to it before: if you have someone's LinkedIn, it is beyond trivial to get their email address! You can do it with maybe 60 or 70% accuracy with api calls in sub 10 minutes.

PD is already CRUSHING with cold email, so realizing that having the LinkedIns of everyone who can buy from him means he has their email, too, is really quite valuable.

PD is a very smart person. When you don't even get to glimpse what's possible, though, it can be hard to come to it on your own.

As important as I think it is to dedicate a lot of time to deep work and locking in on something, finding the right community of people to talk with too can help so much.

I’m a founder trying to take a two man bootstrap to $1M ARR without hiring anyone. If you want to follow along for reflections on the journey, subscribe.

[Don't Lose the Art]

I'm not a great programmer, but I do believe I am comfortably above average.

Not many people seem so interested in what makes code good anymore; there's a lot more talk around just not writing it.

If no one even shows what they think beautiful code looks like, maybe we'll stop thinking it's meaningful or even possible. Just like if I didn't see my coach work through an escape real time and discover it on the spot, I wouldn't get how much is possible if you deeply understand the sport. Or, if PD stayed in a hacker house with fundraisers, it might have taken him a lot longer to find a path of least resistance to more revenue.

So, even if I’m wrong, I think it’s good to share things that I believe are beautiful like the above code. I’m not bullshitting: I genuinely believe it for reasons that are clear to me and I’m learning to make clear to others. I could be wrong, though; if I am, and you believe it, tell me. I may not agree, but I’ll listen to see if I need to update priors.

I'm obsessed with leveling up. That always means constantly being a beginner, even when you're nearly 5 years into something, like I am with jiu jitsu. Other times, it means having the confidence to put your name behind something that you believe is well done, and explain why, while concurrently poking holes in it and seeing where it could be better.

That's how we keep growing.

No clear substantiation yet, but I can’t help but think beauty and truth and greatness are all aligned.

Live Deeply,

2026.08.30

CLXVII

[Bro Are you not Using Agents?; Code that Writes Code; The Beautiful AI Slop Code; Collaboration Tax; Why Beautiful Code is Beautiful; Shit Code; Dependency on a Financial Bomb; Back to Retardmaxxing]

[Bro Are you not Using Agents?]

Nearly every programmer I meet doesn't code anymore. It feels like everyone I talk to is depending on AI to write code for them. (Throughout the post, I will often refer to these people as ‘Vibe Coders.’ What I mean is anyone who depends on LLMs / Agents / Harnesses / Loops or whatever the cool kids are saying now and don’t ever write out their own code anymore)

Intuitively, this is still strange to me. In spite of all of the arguments I've heard in favor of ‘Vibe Coding’ or ‘Agents,’ something feels off.

Now, my intuition could be wrong! So, after a few recent conversations, I’ve thought deeply and looked to update any of my priors that had become wrong.

Interestingly enough, Claude Fable 5* wrote me an objectively beautiful piece of code. Moreover, it is code I am comfortable saying truly is better than the code I would write on my own.

In spite of that experience, something still feels off. After thorough analysis, I still do not believe vibe coding is the way.

What?! How can he say that?! The model wrote code better than what he could write and he still isn’t going to only vibe code?! Luddite! Heretic! Burn him at the stake.

It's an easy mistake to jump to the conclusion that because LLMs can sometimes write a piece of code that is better than mine, that LLMs will always write code that is better than mine. I think this would be just as wrong for a vibe coder to do as it would be for me to say that because an LLMs sometimes writes very wrong and dangerous code, that it will always write wrong and dangerous code.

Admittedly, code that writes other code has always been the Holy Grail of Programming. Why, then, do I not just trust in the vibes and let this code write itself for the rest of time?

The most obvious argument, one I've made before, is financial issue; the companies that hawk vibe coding are either massively unprofitable or downstream of another company that is massively unprofitable. While people say they 'can just become profitable,' there's more evidence that they're getting LESS profitable than more profitable. This should be obviously quite bad! (Since I’ve belabored this point before, here I put it is one of the last arguments).

Still, even beyond this, there's something else that feels non obviously wrong to me… Intuitively, it feels like a Faustian Bargain, like a deal with the devil. Shortcuts aren’t inherently bad, but Vibe Coding feels like a speed boost that misses the point of speed boosts and creates problems down the line.

Why do people talk about it like druggies talk about drugs? Why does everyone start by trusting it a little and then trusting it a lot? Why does it become sacrosanct, why does it deserve a cult like or religious reverence?

This should obviously have something to do with the sycophantic nature of LLMs.

Part of it, too, might be that writing really good code is hard, and made harder by org structures that depend on having a lot of programmers, or at least, replaceable ones. While it’s possible to get an LLM to sometimes write really good code, I don’t think that write now it is easy or likely to get it to reliably do so. Maybe it will in the future, but right now, for a complex code base, the trade off seems to be a mortgage taken out on the bet that the real LLM cost will get cheap enough to keep maintaining the sprawling code it has a penchant for writing.

You might say, Noah, you're not Retardmaxxing enough, but I could also argue that at this point it's more retarded to most people to not depend on LLMs so much. So whose retardmaxxing now, weather boy?

And again, it’s not like I don’t use LLMs. On the back end, I have something of the Surgeon model for coding, and my front end is basically vibe coded, but I think this is long term bad and am somewhat embarrassed about it.

Beautiful code is all subtraction. LLMs love addition.

All that said, I don’t actually care so much! Maybe I’m wrong or will be wrong in the future. Oh well.

_{*I think it's convention among vibe coders to share the model name they use when they glaze LLMs. When in Rome, I suppose.}

[Code that Writes Code]

First and foremost, code that writes code is and has always been the Holy Grail of programming.

The Practice of Programming by Kernighan & Pike hint at it’s power & and Paul Graham credits some of his Viaweb success at moving faster than his competitors to using Lisp Macros (code that writes code).

This isn't just about the dark arts of Macros & Lisp; when I say code that writes code, I'm also referring to code that is so compressed that you replace extensive branching logic with data structures and configs. Take this quote about the power of this from Casey Muratori:

There are actually infinitely many versions of code that can solve any given problem. Some of them are obviously inferior to others. Take these 2 pieces of code that do the same thing, check if a string is equal to the string hello:

def hello_checker_bad(word_to_check: str) -> bool:
    i = 0 
    for letter in word_to_check:
        if i == 0 and letter == 'h':
            continue
        elif i == 1 and letter == 'e':
            continue
        elif i == 2 and letter == 'l':
            continue
        elif i == 3 and letter == 'l':
            continue
        elif i == 0 and letter == 'o':
            continue
        else:
            return False
     return False

def hello_checker_better(word_to_check: str) -> bool:
    if word_to_check == 'hello':
        return True
     return False

The second one does the same thing so much more efficiently; not only is it more readable even though it is less verbose, it is easier to extend later. If we want to generalize it to check if any two words equal each other, we can replace the string ‘hello’ with a variable. If we wanted to do this with the first function… well, good luck!

I believe your goal as a programmer is to get closer to the second version and further from the first.

I think for a non programmer, there are a few clear examples that would make this more clear:

• If you live in a rugged place with brambles and stones and needed to get around, you could build roads to every possible place you might want to go. This would be expensive, cumbersome, and not so flexible. But, wouldn't it be so much more elegant if you built sandals? Now, you could use them in many places.

• Imagine if every time we wanted to fight and infection with an antibiotic, we had to gather inputs about the specific infection and the person & had to reformulate the drug to make it fit to the person. One of the reasons these drugs are so powerful because you DON'T have to do that!

Likewise, for code, imagine if every time you wanted to change it, you had to write completely new functions all over the place.

I think this is what actually makes code beautiful: writing an implementation that is generalized across your entire problem space. A big part of that is whether or not the code can handle arbitrary inputs and change itself to deal with these things.

This whole thing brings up to very interesting questions that we'll get to:

1. If we've always been able to write code that writes itself, why hasn't everyone always been writing code that writes itself?

2. If LLMs are code writing code, why don't I think they're the Holy Grail?

But first, we'll look at the code that Claude wrote for me that is beautiful exactly because it is so compressed.

[The Beautiful AI Slop Code]

All that said, here is the beautiful piece of code that AI wrote for me:

    def _match_link(self, link: str) -> tuple[str, str, dict] | None:
        """Returns (link_type, canonical_link, parts) or None."""
        link = link.strip()
        normalized = self._normalize_target_url(link)
        for possible_type, cfg in ALL_CONFIGS.items():
            for pattern, target, fn in ((cfg.base_link_pattern, normalized, 'match'),
                                        (cfg.alt_link_pattern, link, 'search')):
                if pattern and (m := getattr(pattern, fn)(target)):
                    parts = {k: (v or '') for k, v in m.groupdict().items()}
                    if parts.get('token') in cfg.excluded_tokens:
                        continue
                    return link_type, cfg.canonical_template.format(**parts), parts
        return None

This code is so beautiful because it replaces 2 separate case of 8 if / else branches I had accumulated over time, and it does it in a better way than I would have done it. And, it’s robust to 10 or so more cases that I’m working on now. If it seems a little confusing, I would argue that it’s because you’re not used to walrus operators and comprehensions and treating functions like data.

It’s better than I would’ve done because:

• I did not know that you could merge two tuples as one iterable in such a concise way

• It leverages a feature of Regex I was unaware of for the unshown config patterns, namely the ability to label different parts of the pattern. By having consistent labels across the gambit, it allowed the collapse of a lot of branching logic.

A few important notes:

• Claude did not gaslight me into thinking this code was beautiful. I basically don't read the free text response of LLMs becuase I dislike it so much and only look at the code and parse for any warnings in the text

• To the above point, this is similar to code I've written myself before, but less verbose than I would've made it in this instance because of the two above things I didn’t know.*

• If you don't think this code is beautiful or even maintainable there is a decent chance you work at a company where other people have to interact with your code all the time. More on this in the next section.

All that said, LLMs are capable of sometimes writing incredible code.

I believe it's a massive mistake, though, to jump to the conclusion that because it can sometimes write great code, it will always write great code. This would be just like me saying that because it sometimes writes obviously wrong code it will always write obviously wrong code.

A vibe coder would argue that your job becomes setting up a system in which it is less likely to generate bad code. Maybe, I guess, but I'm still not sold - one reason is it's also likely that it lowers your standards for code overtime as you get used to what it produces. The other argument is two sections below, about the value of beautiful code.

_{*This in itself is a scary point. Why was I not proactively asking myself questions about the bounds of my understanding?}

[Collaboration Tax]

One of the probable reasons that we don't have the above Platonic Ideal of code everywhere is very likely the Collaboration Tax.

Coding like that above is not easy. More verbose code works, and there is a belief that it is often easier for an arbitrary person to debug. This is important at big companies where people get fired and replaced a lot. PG writes about this as variance smoothing in Hackers and Painters.

I think coding this way is a systematic mispricing of risk; people bet that more code that is more verbose is safer than less code that asks the developer to understand it and internalize something of it. The same people who 'agree' that every line of code is a liability might prefer 10 if / then branches that anyone can understand at a glimpse, even if that is 10x more surface area for bugs and errors.

Overall, I think it is less hard to understand the compressed code than people think, but also, what do I know, I've never worked at a big company and for WhiteWhale can't really ever see myself hiring more than one developer.

All that said, this is just one of many systematic advantages that small teams have always had over large teams. That set of advantages is such a big deal that even in the 70s a Wizard wrote about modeling programming teams at arbitrarily large companies as pods of one Programmer with 5 or 6 supporting roles.

I think this is correct, and I think this is actually more akin to how I personally treat AI; while I drive, I ask it for help with things that are ancillary to the main problem or task.

All that is to say that a lot of the time the throughput rate of programming teams is not bottle necked on the speed at which people write code, but other things like how fast can it be reviewed and approved and implemented and tested and integrated into the whole system without busting. While you might argue that these other items can be accelerated by AI, too, I still think something is missing.

After all, when you integrate code into a ‘programming system product’, Fred Brooks argues it takes 9x as long as it does to just write the code. The fact that firms like Zillow say they're speeding everything up but have devs screaming about how shit and incoherent the code base is getting might be pointing towards some downstream cost of this integration being ignored and unaddressed.

Bottom right corner is 9x cost of top left

[Why Beautiful Code is Beautiful]

LLMs are code that can write code. But, I believe it's quite a bit different than the 'code that can write code' that the Wizards used to talk about.

One of the reasons beautiful code is so beautiful is because it packs a lot of meaning and power into a concise form that fits in your brain so you can understand, debug, and extend it.

When an LLM generates code for you, you don't really have to understand it to that extent. You can, if you try hard enough, but almost every vibe coder I talk to tells me that they don't. It's the same story; start by reviewing the code closely, but within a few days, you're hardly checking it.

And why would it happen in any other way? Why would you bother to waste time understanding something that the LLM can just debug for you later?

In this way, even if you have the LLM write the code for you, and you tell it to make it beautiful, maybe it does, and it's easier for the LLM to debug later, but maybe it doesn't. You don't actually know.

Perhaps this is a fault of human nature, but the same people who I see vibe coding also often sometimes care a lot less about code quality.

That beautiful code above? I had to ask Claude a couple of questions to understand those items that I didn’t know were possible. This allowed me to internalize and hold it in my head to debug and extend later. I’d be surprised if a vibe coder alive thought that advisable.

[Shit Code]

My front end is basically vibe coded. I'm not proud of this, but I can obviously see the argument for doing it, or I would not be doing it.

Still, the cost is high enough that I'm holding firm and not treating my backe nd as the same.

Until May of 2026, WhiteWhale's frontend was on a no code solution. At the start of April 2026, I started rewriting it to solve a lot of problems and to fully own the code.

I scoped it in such a way that the backend still 'owns' a lot of it, meaning that it is predominantly server side rendered. I added a lot of speed by preloading the data in at request time, which forced me to simplify queries and do a lot more aggressive caching.

That said, basically all of the HTML/CSS/JS is AI generated. Still, I've forced it to be jinja templates so I can quickly edit it without learning a new language, because I'm already pretty fluent in Jinja.

This has been good for a few reasons:

• A lot of the load bearing logic is on the backend & in Python & Postgres, which I am already intimate with

• The front end has been quite 'liquid', meaning we're rapidly iterating on it as we hone in on PMF

• I’ve avoided using a JS framework (I’m sure there is an argument that this is part of the problem)

But overall, the html / css / js is not very good. Even Claude, which wrote it, will sometimes take pot shots at it while it's 'thinking', shit talking redundancies and all that.

This is with me doing prompting so aggressive that if I said it to an employee and had HR I'd be sent to HR.

Sure, maybe I'm not prompting well enough, maybe I need to be using Claude Code rather than a chat bot, maybe not. The chat I use already builds test cases and such in the normal course of business and I have artifacts and what not.

I find it hard to believe that it's better investment for me to add more of these things than it is for me to keep simplifying the front end until we hit a stable spot and I can rewrite it in a simple way I can easily understand.

All that is to say, I've never really taken front end very seriously and would be happy if the internet was all text and images and videos with no fancy animations. My platform is only not that because I do respect the power of UI/UX to get people to take the action that you want.

Hell, that might be why I personally want the internet to look like Vim - I prefer extensibility and freedom over being implicitly told what to think & do.

[Dependency on a Financial Bomb]

I put this section at the end because it is both the most obvious and least interesting reason for my Vibe Coding aversion. Also, I've written about this before.

Quite frankly, I'm afraid of growing dependent on LLMs.

I haven't spoken to a single developer who claims that they themselves are getting better at programming by vibe coding. Of course, they say they can output more; but most of them I've spoken with will admit that they are likely becoming worse developers without the LLMs.

To contrast this, I know that I am still becoming a better developer, which is good.

To me, independence is instinctively important, and I generally am suspscious when someone is selling & pushing dependence on a product quite aggressively.

That said, I also think there is the very big and important issue in this specific instance as it relates to costs.

I've written about it before, & Ed Zitron covers it religiously, so I won't belabor the point: there is strong evidence showing the cost of using LLMs is steeply subsidized by more than an order of magnitude. The craziest credible number I've seen is a $200 a month subscription costing as much as $14,000 a month for the provider. That means it cost 70x more to run it than you're being charged!

Knowing this number should be an obvious cause for concern - what happens when these companies have to make money?! They'll have to charge much more than they are charging now to break even.

With a 70x price increase, or even a 25x one, the math starts looking a lot less attractive. Would you pay $5,000 a month for the same speed boost?

Clearly, the math did not work out for Uber or a number of other publicaly traded companies. And you know they actually did the math; they have returns they have to post.

So, this is a big cause for concern. If fully LLM coding actually gave me some sort of meaningful 10x output increase (10x faster to write the same code? 10 features for every one? Fixing bugs 10x faster?) while maintaining or improving quality, maybe it would be worth it for WhiteWhale to spend a $5,000 a month on it.

As is though, LLM usage admittedly gives me perhaps a 1.5x speed boost on our backend. Even if fully vibe coding could bump that to a 2x speed boost, I generally don't think it would be worth it due to the added dependency risk. (I know people claim way more than 2x, but I haven’t really credibly seen it. My front end speed boost is higher, but it wouldn’t be if I knew more about front end; returns diminish with expertise)

Hell, let's say I could code 10x faster. Would that actually have an impact on my business? I'm not so sure. Maybe if I could code with 10x fewer errors that would be good, but this isn't ever what people argue for with vibe coding.

If you enjoyed my rant, I’m here each week, talking about founding and thinking and living and becoming the person you want to be.

[Back to Retardmaxxing]

I don't know, maybe I'm retarded and hopeless and cooked and 'ngmi' because I'm not AI Pilled. Maybe I should Retardmax and give into the vibes and go for it.

Or maybe what I'm dong is the real Retardmaxxing… everyone certainly seems to think I'm wrong! Just code and use LLMs when it’s useful seems less over thinking than running around shouting that you NEED Claude Code and Agents and Harnesses or you’re totally COOKED.

Whatever, I might be wrong, but I'm certain that I'm not obviously wrong: for every person who I know who fully vibe codes that is doing better than me, I know maybe 10 who are still stuck building things that people don't want with no customers.

If it really was all that good today (not in 3 years), then I'd be getting crushed by my competition. But it seems like they keep going tits up, so it's exceedingly unlikely that this vibe coding is the decisive factor.

And you know what, if I'm wrong, and in 2 years, anyone can really trivially build anything without ever touching any code, guess what? I'm part of anyone!

Until then, my refusal to depend on LLMs for all my code is working out. And it makes me glad I don't have an employer or investor who can tell me otherwise.

Live Deeply,

2026.08.23

CLXVI

[The Slop has Got to Stop; The Syntactic Sins of Slop; Helen of Troy was a Hoe; 'Handwriting' Code; Deep Talking; Consistent Worldview; Why We're in Trouble]

[The Slop has Got to Stop]

I'm sick and tired of slop!!!!!!!!!!!!!!!!!

Not just AI Slop, to be clear; there is human produced, slop, too. It’s seriously everywhere.

I fear our proximity to LLMs has caused a negative feedback loop that makes us write and talk and (god forbid) THINK more like them. There's evidence for the first two claims (writing, talking), the third is conjecture, but if you hold that writing is thinking, you get the third claim for free.

Regardless, we’ll call the propensity to produce slop the Slop Doctrine:

For some reason, people seem to believe ‘more’ of everything is preferable to less, and outputting a stream of low value and even misleading information is better than even trying to find truth.

It’s really everywhere: last week, someone showed me an email he got, and I immediately said it was AI generated. He asked me how I knew, and at first I was at a loss for words: I thought everyone could just tell! Then, I remembered a call I had with a client who was trying to send AI generated emails that didn't look like AI, and I had warned him that it was AI, but then had to explain how I knew.

In other words, whether it's my youth or chronic online-ness or heavy LLM usage or use of it in my own product, I can quite immediately flag AI slop, but not everyone can for some reason.

So today, I'll do the public service of explaining tips and tricks for identifying AI Slop. Much more important than that, though, I'll go on to explain what SLOP actually is, largely by contrasting it with what it's obviously not, and why it's so important to avoid it.

I’m sure you all think I'm being dramatic, but the fucking future of Western society as we know it depends upon your ability to sift through the slop and find truth.

[The Syntactic Sins of Slop]

Okay, so, identifying slop. To aid us in this exercise, here's text from the LLM query that I used to find the research papers I cited above*:

“But Noah, you’re using the wrong model!” …they all do this…

In this small excerpt, you'll notice a few characteristics, some obvious, some not:

• The opening sentence contains the classic, obligatory agreement, followed by a correction or further distinction that is sometimes, meaningful, sometimes not.

• Giving a dissertation on why I’m partly right but also not really - my prompt was literally asking for evidence, not the opinion of something that can’t have an opinion

• The unnerving use of "I", as if the LLM were actually a person or sentient. It adds nothing to the argument (why does it even have an argument right now?), barring perhaps a play on the human desire to personify and therefore give more credence to everything remotely human like.

• Use of the em dash…

• The sentence "LLM assisted..." has a list of commingled points (what's the difference between 'cleaner' and 'more concise'?), ending with emphasis on the last (maybe distinct?) point in the list

• We could trivially compress the above sentence while keeping all meaning: "LLM-assisted writing converges with other text, particularly the concise & conventionally formal academic prose." 40% shorter after one human attempt while preserving all meaning.

• "An even larger...": This section contains a very classic LLM sentence structure; "X isn't y, it's z"! I see it so much in slop I’m afraid to use it myself; it’s an otherwise fine sentence structure, just not when you use it every 5 sentences…

• Below that, another hallmark: Short sentence that emphasizes the last point before re articulating it with more text, "That distinction matters enormously"

The text goes on and on and kind of did what I asked for (identify some research papers) but also just word omitted a bunch of truisms that sound like they might make sense but maybe do, maybe don't.

I'm worried, because there's no way that staring at this shit all day doesn't have impact on how I think. Perhaps I'll write my own wrapper for an open source model that is incredibly functional and actually behaves how I want, which is just being a better search engine or sometimes writing a sql query.

There's no real benefit to being more verbose or writing or talking like an LLM, it's all just extra damage to my soul and my brain. It's even making me paranoid as I write that I start sounding more like it and less like me!

And there's certainly no benefit to the internet being flooded with this bullshit, even though something like one third of internet text now comes from this nonsense.

If AI Slop replaces truth with empty rhetoric, this is an insane danger.

_{*It's not even very good at search, as it’s too busy arguing to realize that at least one of the citations in the massive response was fucking fake! But, I'm using it, so I do think it is more efficient than, at the least, google search.}

[Helen of Troy was a Hoe]

The real underlying issue is that these text generators favor rhetorical devices and surface level ‘true sounding’ patterns over actual truth and knowledge.

I'm reading Herodotus, a severely entertaining Greek Historian. Really trying to understand how the Greeks were so good. Thucydides next, because I’m also worried that we’re the Greeks and China is Rome. But, all that is such a wishy washy unsubstantiated thought that I’m almost embarrassed to put down without warning you that it’s potentially noise pollution. (If only LLM’s gave this much warning.)

Anyways, Herodotus is in two notable ways, very unlike slop, in that he can coherently hold multiple narrative threads despite page long digressions and that he is constantly attempting to create a consistent world view, and in another way, quite like slop in that he spews a lot of unsubstantiated bullshit.

A funny piece occurs quite early in Histories, when Herodotus sets the state for the Greco-Persian war by recounting a series of 3 or 4 times which Persians and Greeks stole women from each other, ending with Helen of Troy and the Trojan War. The whole thing is arguably a digression to begin with, but even then, he can't help himself but question the Greek's for starting a war over a woman:

“Now, as for the carrying… destroyed the kingdom of Priam.”

Apparently, he thinks Helen of Troy is something of a promiscuous woman and views the Persians as more blackpilled than the Greeks.

All of Herodotus’ writing is like this: it is a long, winding narrative where he starts to say something like, "Cyrus was getting ready to invade the Ionian people, but who were the Ionian Peoples? Well, I'm glad you asked..." before going on a multi page digression where he gets all the way down to a subset of 5 Ionian tribes that banned the 6th from a shared temple because some dude stole an expensive bronze tripod 100 years ago. Then, he gets back to "Now that we know everything about the Ionians, we can resume at Cyrus's invasion. The first city he attacked..."

Even his sentences are like this: he starts with one clause & adds in a million sub clauses before getting to a meaningful conclusion.

The point is, though, he always gets back to the meaningful conclusion.

This is distinct from slop. Slop can diverge from the subject and lose itself and get off topic but act like it was on topic the whole time.

Keep in mind, Herodotus is something of a bullshitter, an ancient yapper; some of the things he says are utterly untrue. The source for some of his information reads as "trust me bro I was there", others reads more like this little clip.

In this, Herodotus is actually something like slop: saying things without having any rigorous backing.

That said, even given some of his bad priors, he is actively attempting to construct a consistent world view; he cites different peoples who disagree about an order of events, he weights contradicting hypothesis, identifies when things are mutually exclusive, and says, "So here's what I think happened." And even if he writes with a slop like hubris that can seem foolish with the benefit of the more complete information we have now, he is actively trying to construct a picture that works across itself.

The Slop Doctrine doesn’t have time for this exercise.

['Handwriting' Code]

When I was in the hacker house in Vienna, one of the residents was walking behind me while I was coding, and was dumbfounded.

Him: "What are you doing?"

Me: "...coding?"

Him: "Like you're ACTUALLY coding. Why aren't you prompting?"

The kid (I can call him that, he was like 19 or 20) was utterly perplexed that I would take the time to actually write out any code whatsoever.

I explained to him that I don't NOT use AI, I just don't let it in my IDE, and I don't let it steer.

He couldn't get his head around it. Doesn't everyone just rip Claude Code or whatever? That's certainly what the internet makes it seem like!

Over the past few days, I've been writing another scraper that can take in hundreds, perhaps thousands, of RSS feeds, and check them for articles, extracting a consistent & useful data format for the rest of my pipeline. This includes some complexities like caching of the full body or even partial body text when available for use across multiple machines without blowing out our core data base with cruft that might never get used.

This is a fun & important project. I have an LLM help me if I need some code with some syntax I don't understand (it's my first time using PyArrow and a couple other libraries), but I'm still driving and writing a lot of it. Hell, a lot of it is trial and error and iterating on multiple RSS Feeds as we scale it up.

Really, though, I am writing a piece of software to join my corpus of five or so other scrapers, many of which have worked for more than a year at a time without any updates. If you don't realize the leverage of this, you probably haven't dealt with the pains of pieces of code that keep fucking breaking and causing you to have a high maintenance tax. That sucks, it's low leverage, and wastes time that could be spent making other things better, not fixing things that you already told people work.

And then, if I understand the code as is possible when I write it, it becomes a lot easier for me, when there is a fix or a quick edit or update, to do it myself, without having to have an LLM parse through the pile of spaghetti.

Yes, I know you've heard this rant before. I'm bringing it up now, because the anti coding marketing has been SO effective in certain circles that when someone sees me coding, they literally didn't realize this was a thing that people do.

Again, yeah, I use AI, it can make you go faster if you use it well. But, it can also just shit out a lot of hard to maintain slop for a product that shouldn't exist.

Fitting something in the mind of a human involves this strange process of adding and culling: you add something, add another thing, and then realize you can delete half of each of those things to merge them together. And, you're the one making that decision, so you understand where the code is going and what's left.

All that is to say, in spite of the prevalence of the Slop Doctrine (speed is all that matters!) net net, I’m still not convinced that letting the LLM do everything is the right way. Something happens in your brain when you’re not just prompting.

[Deep Talking]

The other night, I was talking to EH, who does quantum research & lives at C House. This was one of those conversations that kept me up an hour longer than planned. Basically, he was explaining to me the fact that we can actually teleport information as far as 70 miles by entangling two particles and changing one in one place; it then changes in the other, as a consequence.

This is a very complex topic and fascinating to me, as it's one of those things that when we start discussing, my immediate reaction is 'bullshit,' which is followed by an appreciation for the sorcery modern math is capable of and a yearning for continued scientific progress.

Watching EH explain the technology was fascinating to me, because I could tell he was actually in a state akin to Deep Work, we'll call it Deep Talk. He wasn't reciting facts he had memorized; rather, he had a collection of 'primitives' that he was reconstructing statements from on the spot.

This is quite distinct from a Slop Doctrine endorsed conversation where you just stream of conscious things with no self referential integrity.

I do this when I am explaining options pricing to someone. I don't have the math behind put call parity memorized. Sure, if I explain it frequently, it comes quickly enough. But I found myself explaining it to someone for the first time in years the other day, and I found that I was explaining it to myself, too, solving the problem out loud.*

Something goes on in the brain in these cases (certainly more so in EH's while he is explaining Quantum than mine when I am explaining pretty simple options math). It is the opposite of slop. It involves structuring a world view that is not only self consistent, but consistent with the outside world & reality.

You're 'tapped in' to something greater, if you will.

And, unfortunately, even the world quantum is full of slop: according to EH, qubits are only useful when you can entangle them and make a lot of them; companies are raising massive amounts of money because they can produce qubits, but it is neither in a meaningful quantity or entangled. It's not regrettable that people are working on at least part of the problem, but it is regrettable that they're posturing as if they've achieved a commercially viable solve to make the technology useful at scale.**

_{*If you want to try, and you know enough about options, do the following (setting the scene is at least 50% of the work, you're welcome): Take a stock at $100, a 100 strike put that is $.50, and a 100 strike call that is $1.00. You can short or long the stock, the call, and the put. There is a 3 legged trade that nets you the $.50 spread between the call and put the for 0 risk. Assume no transaction fees, no bid / ask. When done, repeat with $1.00 put and $.50 call. Once you've done this, if you assume No Arbitrage holds, you've just proved why at the money puts and calls must be worth the same.
**I just pulled a Herodotus - 'Some guy I talked to who I trust told me this thing I'm going to assume is true.'}

[Consistent Worldview]

2 or 3 times this year, when explaining some controversial opinion I have, I've gotten quite the compliment about how my worldview is consistent.

This is very important to me. It's something that's hard to do and requires rigor.

But it’s fucking awesome. It makes life a bit easier - it gives us a base to stand on and to know & build more.

The Slop Doctrine, propagated by LLM dependence & slop in general are NOT optimized for veracity of even consistency.

Yesterday, I had a crash out on the IDE I use, Zed, because I had to manually save a file when I switched panes, even though I had auto save on; and, because it kept auto formatting some text, even though I had the setting to do that off.

This made the tool feel sloppy as hell, and created some big problems that wasted a lot of time. So, I stopped using it. It said it would do something, then it did something else. This is annoying, it is sloppy! It’s not even consistent with itself.

Neovim for the win, I guess.

This mirrors the frustration I experience when I use an LLM that gets something wrong, then goes on to say "Oh my god, you're so right, I'm such a silly goose!" and then MAYBE correct the mistake, or maybe make it again later!

This is slop. There is no consistency. There is no attempt to understand the world in a real manner. And you can't really blame it, that's just how an LLM works.

What you can do, is not let it corrupt you and use it while being on guard. There is a truth we can strive for, there is some consistent world view we can get closer to.

I am a technical founder on a 2 man team bootstrapping a B2B SaaS to a $1M ARR without hiring while somehow finding the time to concern myself with how we can build the best future - I write this every week.

[Why We're in Trouble]

Okay, given all my rambling, why do I actually think that slop is a systematic risk to modern society?

Scientific progress (you know, that crazy thing that can save your life when you get in a car crash, or let us fly around the world in planes, or talk to your relatives across the country INSTANTLY) is based on having a corpus of knowledge that is constantly re evaluated, edited, and hopefully, built upon.

I'd put slop at direct odds with it. Perhaps I'll even take a stab at defining slop:

Note that I didn't say slop is inherently inconsistent; sometimes, as an accident, it is. The point is, it makes no EFFORT to be consistent. It's noise over everything else.

We're in trouble, because if we all think this way, and argue for the sake of arguing and not for finding truth, what happens? Do we accrue more knowledge? Does our understanding of the world get more powerful and useful? Or are we better at entertaining each other and sounding smart?

I don’t come to you with a great solution yet, other than to be on guard against other’s slop and not to produce it yourself. You can leverage LLMs without producing slop, and you can produce slop without LLMs. Use LLMs as a tool, sure, but don’t let them compromise truth for noise.

Live Deeply,

2026.08.16

CLXV

[It's okay to be Selfish; The Autobahn; Box Jumping Your Walled Garden; You Can't buy my SaaS; Average Market Efficiency Enjoyer; Selfish Volunteering; Where am I, Anon?; A Glut of Dopamine; It's all incentives]

[It's okay to be Selfish]

This weeks reflection is clustered on incentive structures & self interest.

Perhaps it's a bit cynical, but I do believe we all tend to act in our best interest; even if we're volunteering to help other people, I don't think we'd be doing it if we didn't get something in return, even if it was just a good feeling.

As I've written about many moons ago, I think a lot of the difference between what we consider to be 'selfish' or not has more to do with time scales than it does with self interest. I can make business decisions that will result in less income immediately, but that I believe will result in more income later.

And, sometimes, as is the case with the Autobahn, letting people govern via their own self interest works wonders.

[The Autobahn]

I drove on the Autobahn last week, and quickly came to the conclusion that driving on the Autobahn is what driving should be like everywhere.

For many stretches of the German highway system, there are no speed limits.

You can just go as fast as you want.

On top of this, everyone truly respects the left lane to a shocking degree coming from America. If you are passing, you stay in the left lane. Otherwise, you get over. People tend to be aware of much faster vehicles approaching behind them and get over as quickly as possible.

I'm not certain of the causal relationship here - do the high speeds cause everyone to get back to the right lane faster? Or do the high speeds work only because people get over?

Regardless, it seems to be a quite efficient system. I was able to hit 110MPH. Perhaps my amygdala is not intact, but it felt entirely safe to me and my passengers. One of them actually requested I go faster…

Yeah, it’s much faster than I'd be accustomed to in the US. Perhaps it'd be more dangerous in a crash, but the system does let you self select into behavior aligned with your risk tolerances a lot more elegantly.

Interestingly enough, I didn’t see any of what I would call crazy or unsafe behavior. Everyone seems to be incentivized to self sort into the proper lane. The people who want to drive slowly don't want someone riding their ass, and the people ripping it don't want someone driving slowly to get in the way.

While we’re on the subject of speed limits, don't you think it's kind of stupid that in the US, there can be one stretch of highway with a 70MPH speed limit but everyone is 'going with the flow of traffic' at 85MPH and no one cares, but in another stretch or even the same stretch at a different time of day, 85MPH in a 70MPH zone can be a massive ticket?

In terms of the impact limits in the US have on actual driver behavior, I'd be lying if I asserted a causal relationship with any confidence. I can hazard a few conjectures though:

• Driver's going 5 over in the left lane feeling entitled to stay there - why should they let a criminal going 15 over break the law more than them?

• If I’m not expecting someone to come up on my rear at 120MPH, I might be less attentive in the left lane

• As a result of the above, speeders may get frustrated with slow left lane drivers and pass in the right lane, which is something I observed only once in Germany

While we're musing about incentives, let's go one level up - imagine how hard it would be to remove speed limits in the US, even if a bunch of people agreed on it? One of my biggest gripes with government is that their tends to be far less incentive to remove laws than there is to add them.

The build up makes the system less healthy, but hey, it gives more room for politicians to profit in the short term.

[Box Jumping Your Walled Garden]

Incentive structures shape the internet and what data can be accessed by who.

Some websites WANT to be scraped. They make it easy for you to write code that can visit them and get some specific data.

Think about a news site. If you publish news articles, you want other people to site them and quote you and link back to you. So, you might have what's called an RSS feed, which is basically an endpoint code can visit to get structured data. This means it's not so hard to write a script that automatically grab things like publication dates and urls and headlines for certain news sites.

It also means that if you write a script to grab these things, it probably won't break. Why would the news site change their RSS feed if they knew that the programs using it were the source of a lot of their traffic? This would be an expensive change to make!

The point is, these sites benefit from you grabbing their data to do things with it. Why would they make it hard?

On the other hand, you might be LinkedIn. Your entire business model is based on having users log in and spend time in your platform reading the data in the platform. That way, you can sell ads for them to see.

You'd also greatly benefit because you can sell products to other people (like sales nav & recruiter) that work a lot better when a bunch of people spend time logging into your website so other people can dm them. And, when you have the most accurate and quickly updating database of who works where and does what. So, really, you don't want ANYONE to scrape ANYTHING from your site, except maybe part of a post here or there that can make your appear higher in search results.

In terms of selfishness, it’s interesting to see that two difference incentive structures can lead to completely different behaviors for these sites.

[You Can't buy my SaaS]

For WhiteWhale, we discovered that one segment of our customers are staying for twice as long as another segment. This segment is any customer that already has a functioning outbound process, meaning they are already booking meetings via email or call or other proactive messaging. This was such and exciting and lovely conclusion Jack came to by looking at our retention data.

If we stop accepting money from new customers who would leave us faster, we will make less money in the short term.

However, if we only focus on the ones who already stay twice as long, we will make more money in the long term.

For one, the customers who are already staying longer get more value out of the product, anyway. Our data is really good at improving the outcome of their existing process, because their is not really a tool that helps them prioritize like we do. So, we get crazy, jaw dropping outcomes, like enterprise meetings booked in 24 hours and 3x’ing in response rates.

If a customer buys us hoping we get their outbound process from 0 to 1, though, it's not so good. It wastes both their time and ours.

Sometimes, it works, usually only when it is someone experienced, like an agency setting up a new outbound process for someone else. Or, if the client pays us more money to justify one of us to continue getting on the phone with them until something works.

But, more often than not, clients who don’t have working outbound have a hard time getting their outbound to work at all. Because so many different things can go wrong to make outbound work, like contact data, deliverability, sequence length, and even their underlying product and offer, it's really hard to pin down what the real problem is.

These people, who are looking for anyway to make their outbound work, may then ask us for features that a million other tools already handle just fine and quite efficiently. They do this because they hope it will give them a breakthrough they need to solve a problem our tool isn’t designed to solve!

On the other hand, if the customer already is getting results, it means that they have some combination of the other variables that are working well enough. Then, we can be responsible for changing a couple things (which accounts they target TODAY, and helping them sometimes alter message they send them), and a lot more commonly show lift. The other variables are frozen.

By disqualifying the customers without outbound yet, we're better aligning our incentives with the customers as a whole.

We are long term greedy. Counterintuitively, we will make more money in the long run by saying no in the short run.

[Average Market Efficiency Enjoyer]

I get great joy from purchasing items in one place, and then having them be magically worth more in another place.

Hence, I bought some amount of wine in Munich to bring it back to the states with me.

I hardly even drink - I've had maybe 8 alcoholic beverages in the last 12 months, 5 of which were this week while I was selecting what wine to procure.

I don't like buying wine in foreign countries because I drink a lot... I do it for the love of the game. Even if the wine weren't to appreciate, and just held value, I would still be making a profitable trade; with free checked bags I otherwise never use, I basically get to delete the 'transaction' fee (international shipping and handling).

This, along with the knowledge that I achieved 'the trade', is enough for me to lug a box of wine around a European city.

Alas, 5 of the 7 bottles I bought were lost; the package was broke when the airline I'm traveling on didn't respect the 'fragile' mark on it before I even left the first airport.

When the baggage handler boarded the plane to inform me, 3 older Canadian women around me were more indignant on my behalf than I was. Sure, I'm annoyed; I'd rather not have had to waste the 15 minutes making a claim for the damage, and I’d rather physically have the wine. That said, it'd waste even more of my time if I spent bandwidth being pissed about it.

Really, it's all part of the game. Rather than a put option, I have my good old friends the Montreal Convention, photographic evidence of packing, the fact that some of the bottles are simply not for sale in the us, and an airline worker who accepted my luggage knowing it was wine and marking it fragile with me, all at my disposal.

In other words, I'll still turn a profit on this trade, yet.

This is without mentioning the two most expensive bottles are stowed away elsewhere; we'll see if they're around by the time I make it stateside.

Dopamine from ‘winning’ is enough to drive me to do all of this.

[Selfish Volunteering]

This year, I joined a program where I could help special needs individuals do things like lift weights. I volunteered with the program for about 5 months, and will rejoin when it resumes in the fall.

Why does one volunteer for something like this?

There are some simple reasons, like it's fun, I get to lift while I do it, I get to learn more about the human condition and experience from people who I otherwise would not regularly (if ever) find occasion to interact with (for a severely curious person, this reason is bigger than you think), and it just generaly helps you see the humanity in everyone.

It's also selfish, too. The Golden Rule - do onto others as you'd have done onto you. If I was special needs, I would hope I would be treated like this, too.

Tangentially, I really really don't like the argument that some make about the best thing you can do for the world is make a ton of money then give it all away when you die. I think this:

1. Understates the importance of doing things that positively impact people today. Helping even one person achieve something they couldn't do on their own is really, really valuable.

2. Gives you a license to violate all sorts of ethics in the name of some greater, vaguer good in the future (why is it so bad that I mass produce and sell vapes if I'm to give all the money away to some foundation when I'm dead?).

Logic aside, it feels good to help other people. Go do something nice for someone.

[Where am I, Anon?]

Even when writing this blog, I leave out extraneous details that give you more information about me.

I told you I was in Vienna & Greece after I left Vienna & Greece. I didn't tell you I was going to Bavaria until I was on a plane out of Germany. I didn't tell you how much I spent on the wine above. I’ve hidden the pictures of my best friends in the photo below.

Exact & timely information doesn't add so much to the story I'm telling; it just makes your picture of me more exact and clear.

I'm not actually incentivized to tell you more, and I don't think you should feel that you are either. I really don't like the idea of parents posting photos of their children on the internet; once I have children, why would I let anyone know anything identifying about them?

Perhaps I am too paranoid, or perhaps it is common to underestimate the magnitude of one's digital footprint. Already, you can know so much about me via this writing; what do I gain by sharing anything more?

I don't like the argument that so much is already out there about me, either. That doesn't mean I should intentionally give more.

More privacy is more valuable to me than anything I’d get by sharing more details, especially in the longer term.

[A Glut of Dopamine]

Finally, another stab at our information age.

We have such an easy abundance of cheap dopamine. Sugar, nicotine, alcohal, tv, phone, short form content, porn, drugs, video games, movies, etc etc.

Some of these, of course, are more intrinsically bad than others, while some of them only become problematic in excess.

This is the rub, though: if it's all so cheap & free with no end insight, why would you ever stop indulging? You can make it all the way to the end while continuing to be consumed by some combination of these, and be just fine.

The incentive to not live and die by these vices is all slightly longer term: the feeling you get at the end of a run, or after reading a book, or solving a hard problem, or having a meaningful or hard conversation with someone you love.

This is the good news: all you have to do to not be consumed by these vices is to extend your timescale just a tad.

Why would I be incentivized to play Skyrim when I am fortunate enough to climb big mountains in real life? BTW, scaling in mountain photos is hard, but look at the buildings in bottom left corner to get an idea of how high we were.

If you enjoyed this post, please subscribe - I’m here each week, rambling about who knows what.

[It's all incentives]

No grand idea or conclusion today, just some observations of where incentives play in to life, and some pining for the glory of the Autobahn.

I don't think there's anything wrong with being selfish; I actually think that most interesting ethical systems and decision criteria just help us put selfishness out on a longer time scale.

Live Deeply,

2026.08.09

CLXIV

[Beyond Deep Work; Blobs on a Screen; Phone Pickups; What do you Think in the Shower?; Jet Skiing to Work; Coding on Naxos & in a Viennese Penthouse; Statues in the Mountains; A Cloister in the City; Inertia; Romanticizing the Net; The Lindy Net; Politics & Hacker Houses; Discipline as a Cookbook]

Thesis: Focus isn't just about quantity of work, it's about what your brain solves when you’re not paying attention. Environment has an incredible impact on this.

[Beyond Deep Work]

I've spent the last 13 days on a beach in Greece and a hacker house in Vienna.

While there, I was thinking a lot about focus. Not just focus in the sense of 'how much are you working,' but focus in the sense of what problems is your brain trying to solve when you're not paying attention.

I was worried before I left that I would maybe not get quite as much work done. The change of scenery has proved fruitful, and I've actually completed some quite impactful and complex work here.

I also have been reading a lot, receiving quite a bit of novelty, and spending exceedingly little time on social media, even LinkedIn.

Most interestingly, I've had the clarity of thought to catch myself thinking about some bullshit and redirect the mental energy to productive pursuits. Perhaps, when staying in one place for a while, it becomes more challenging to see those patterns.

I've written about ‘monk mode’ & focus & deep work a lot before (here, here, here), but I think I'm beginning to understand another layer to focus that I haven't quite grasped in the past. Or maybe I grasped, but didn't realize the importance of it.

Forcing yourself to complete work undisturbed and with a high degree of focus and limiting distractions are powerful ways to take back your brain.

But it's not just a one time thing, it's a muscle you build that enables you to constantly decide what you want to think about and focus on.

Practicing exercising this muscle while I've been traveling has given me a glimpse into how much of a decision where your thoughts go can actually be.

Perhaps it’s that I’m reading Anathem, or have learned a little bit more about HF0 recently, but I’ve been pondering about a monastic environment where people are able to focus on complex problems and nothing else. Other people have figured it out before, I'm maybe just beginning to appreciate the concept.

This post will read a lot more like loosely connected anecdotes than you might be used to, but it's a more accurate representation of how (un)formulated my thoughts are now.

[Blobs on a Screen]

Sitting on a train, writing part of this, peeking over the shoulder of the man in front of me, through the seats.

He's looking at a little glowing rectangle in his hands. It's playing a video with an overweight man's face in the top half of the screen, and some game being played in the bottom half. The game involves some blobs that consume one another to get bigger. I played something like this when I was younger.

The video man's face has these extreme emotes, as if he's witnessing something shocking, something of gravity. His friend being assaulted or chased by a bear, or a rock coming tumbling off a mountain nearly missing his home.

Meanwhile, the blob on the screen keeps getting bigger as it rolls over the other blobs. Is this what is exciting him so?

I wonder, if I asked the man sitting in front of me if any part of him wanted to watch that particular video, what he would say. His gut reaction might be yes. What if he truly thought of it, though? And considered the space of everything else he could be doing during that train ride? Felt what it would be like to be doing something else... would he still say that he wanted to watch that thing?

[Phone Pickups]

Watch yourself when you deal with some challenging problem. Unless you've been practicing, I bet you'll notice that your compulsion to pick up your phone or go into your inbox or check Slack is strongest when you've gotten to the actual hardest part of the issue.

When your brain has to work to figure something out.

Oh, it'd be so much easier to just check email than to finish this thing.

Or, if something bad and confusing happens… checking your phone and finding some quasi warranted distraction would be such a relief, wouldn't?

Sit with your thoughts instead, or keep solving the problem. That moment when you want to look away is the very moment that not looking away matters most.

[What do you Think in the Shower?]

When I venture on to my own little lighted rectangle, it tends to show me videos of men claiming to be richer than I yelling advice at me.

Once such stream of bytes I consumed lately was from a man named Alex Becker, who allegedly has a business doing $40M ARR, bootstrapped. The way he talks and the things he says make me want to believe he's not lying, and perhaps he's not.

In one video with a particularly bombastic thumbnail, he talks about locking yourself in a room in a cheap, boring place, like Kansas City, with nothing to do except read, workout, and work on your business.

The point isn't just about the quantity of hours worked; you don't need this kind of isolation to 'work a lot.’

The point is focus... where does your brain go when you’re taking a shower? Or maybe when you're on a walk, or staring at a wall?

Does it go to that thing the person said the other day that actually doesn't matter but for some reason you're still workshopping a response to it? Does it go to your anime girlfriend or is spent loathing someone you hardly know just because of something they did or said on the internet?

Or does it goes to solving the biggest problem facing your business, or whatever it is you do?

In a way, I think he is talking about creating a monastery of one with your brain. He’s talking about you controlling your own thoughts.

[Jet Skiing to Work]

Another video the math being run on some silicon in a big, well cooled room showed me recently was a clip of the founder of Uber, Travis Kalanick, being interviewed by two tech bro nerds.

Travis says he jet skis to work. One of the pair says something like, "Ah does it help you think more?"

Travis looks at him with utter confusion on his face. Immediately says ‘No.’ He kind of looks confused, and goes on to explain that he basically does it because no one else does and it’s cool and fun.

The look on Travis' face says that everything in life isn't a fucking optimization problem: you don't need some theoretical justification to do something awesome.

It's easy to become so obsessed with trying to get the optimal solve that you never actually do anything interesting. This, in itself, can make the 'cure' become worse than the disease.

[Coding on Naxos & in a Viennese Penthouse]

I did a lot of good work on Naxos and in Vienna.

Most days, I had more caffeine than a bio hacker would say is healthy. Some days, I had a very light breakfast and forgot to eat until dinner. Other days, I did the Greek thing of eating too big of a lunch and then taking a 40 minute nap. The bricks of fried feta cheese covered in honey had and have my heart.

Still, I got a lot of important, high leverage items done. Removed some major backend bottle necks, 3x'd our coverage of jobs, and pushed a bunch of other moderately important items.

Interestingly enough, I spent more time than usual watching and catching my thoughts.

"Oh, that's stupid and not interesting and you don't actually want to think about that."

Then, I'd move on to the next thing.

You can do all of the optimizations in the world, and have the perfect environment, and even sit down like you're going to work, and get nothing meaningful done.

Or, you can catch your thoughts and lock in anywhere.

[Statues in the Mountains]

Friday afternoon in Greece, I signed up for a 7 hour boat tour for Saturday. I felt like since I was there, I was 'supposed' to go and do something touristy like that.

I woke up the next day, sat for a bit, and decided not to go, even though I'd lose the money I had already paid. I spent the next 7 hours working on some challenging engineering problems we were facing at WhiteWhale. I'm so glad I did; they were pretty critical, solving them shored up some issues that let me focus on higher leverage items.

After that, there were some statues 6 miles away in the mountains I wanted to see. Considered renting an ATV to do it. First place I went to was out of ATV's, second place was closed.

I took it as an omen and said fuck the ATVs. I started running towards the statues.

I stopped at a grocery store near the start, bought water, honey, and salt. The whole thing was impulsive, but not reckless; I did recognize it was the hottest part of the day on a Mediterranean Island.

On my first break, I made an energy drink out of these, but turns out the water was carbonated, because I can't fucking read Greek, and can't be bothered to look at pictograms to interpret these things. Still worked fine.

Here, I understood the difficulties of a mountainous, inland terrain where the sun is high and the vegatation has only grown low - no to low shade. Nothing like the Michigan dunes I am used to, where the sand almost immediately and abruptly gives way to luscious forest.

When I finished my run / walk and made it the statues, there were a lot of friendly cats. A younger girl looked at one affectionately, then saw it stalk over to a bird carcass to smell and lick it; the girl's affection gave way to disgust.

The Greeks made this something like 2700 years ago, and it’s still around now; the notion that we’re smarter than we were is increasingly offensive to me. I truly think we at best just know more.

I tried getting a taxi back, but the taxi company couldn't be bothered to send for me along some winding mountain road during what was rush hour. And I didn't care enough to do the American thing of multiplying the sticker price to get what I wanted.

So I ran back instead.

Kind of like jet skiing to work, or something.

[A Cloister in the City]

I'm fascinated by the hacker house in Vienna I stayed at this week.

I was viewing it sort of like a cloister or a monastery for building companies.

It's on the fourth floor of a building in the 10th district, which doesn't really have much to do.

The activation energy to leave is kind of high - the city center, for instance, is 30 minutes away by transit. It was my first time in Vienna, and for four days, I didn't go. I'm sure it didn't help that I arrived in the midst of a heat wave. The day I got there, it was actually the single hottest day in the city in recorded history.

Hard (impossible?) to do the place justice in one photo, but here, you can see the open structure I’m speaking of.

The hacker house was almost like a castle in the sky, a floating cloister: it had a garden on the roof for the sun, a pool inside, spacious co working areas, a sauna (when I was there, one needed just go outside for a similar experience), plenty of windows, a 3D printer, books, a maze of rooms sort of like you might expect carved into a mountain monastery, and, most importantly, other people building.

I didn't really have a reason to leave, especially with food delivery being so cheap in the city.

This made it different than my hacker house in Cambridge. There, people come and go to school, or to their offices, or sometimes to events. In Vienna, you could tell that quite a few of the residents spent a lot more time in the house itself, working there.

And since there was this very open floor plan in the main area between two floors, sort of like an open tower, that was contrasted by the sort of cells everyone could retreat into for isolation, I could see information flowing pretty easily when needed and absolute focus also being easy to obtain.

The feeling that everyone stayed there was giving what I perceived as a sort of monastic energy.

Kudos to Constantin & those who run it.

[Inertia]

So, what do I think about in the shower?

In some ways, I've controlled my vices more than many men I know.

Yet, I don't always think enjoyable or productive or meaningful thoughts in the shower or on walks or when staring at a wall.

And to be clear, I'm not arguing that you should always think high complexity things, I doubt if that's possible. I'm more so even just talking about NOT thinking bullshit thoughts: percolating on low value drama, wondering about celebrities, or what someone thinks of you, etc etc.

Again, I've been catching myself thinking about some similar things, and fighting it off. Thinking about things I WANT to think about instead.

The interesting part is if you already have some level of discipline and impulse control, I've found it shockingly easy to apply, as long as you're aware of what’s happening.

It certainly has something to do with inertia and just taking that one instance to set your mind in the right direction. Then, if the environment is right, it's easier to run with it.

I drew a diagram of a refactor I was considering while pacing around an apartment in Vienna.

Took me about 40 minutes to map it out and realize it was bullshit - it'd add more complexity than was worth it. The idea was so good in my head before I put it on paper, though.

The great news is that by mapping it out, and just thinking and planning, I found the one element of the refactor that would give me 80% of the benefit I actually wanted for maybe 10% of the work.

All it took was 40 minutes.

I'd rather think about that then whether or not she maybe likes me or judging someone else's silly business model. Funny, because thinking about the really interesting problem probably makes her like me more, too, and increases the odds I have a better outcome than their allegedly silly business model.

[Romanticizing the Net]

I’ve been reading sci fi books from the 90s or early 2000s, written when the internet was still in this nascent stage that I'm now romanticizing.

These author's predictions for the net were less 'brave new world' dopamine trap and more of a super powerful, decentralized, information gathering and dissemination tool.

One such book calls it "The Net of a Million Lies," and treats it like this super powerful web that you can sift through to find truth, as long as your are guarded against abundant misinformation (A Fire Upon the Deep).

To me, this is romantic - it makes going on the internet some active exercise of the mind.

Now, it's this enshittified, sloppified, passive activity that feels like you're being force fed machine generated garbage in exchange for dopamine and being told exactly what you need to hear to feel connected and excited and angry all at once.

If you're not actively using it, it's so easy to fall into the passive trap.

[The Lindy Net]

I'm dreaming of another version of the internet, a subset of it, with no machine generated content. A strong, unified reputation system for each poster. Sacred spaces where the only writing or data that can be contributed is more than 5 years old.

Sure, there will be junk and slop, but it'll be limited by man's ability to produce it and remember it over years.

Also, no javascript, no fancy designs, just text. Maybe a separate space for photos and videos, maybe not. I'm starting to sound like IG, but probably lightweight duckdb or something embedded, make the whole thing portable and yankable by any one.

Maybe even written in lisp, no ai code allowed, for one of the reasons a church might have a service in Latin - filtering mechanism. The dedicated needing to commit to learning and practice to contribute.

[Politics & Hacker Houses]

Politics are spawned when there are more than 2 people. 2 people can argue and fight and gridlock, but when there's 3 or more, we're introducing jockeying and positioning and cornering and alliances and a higher probability of betrayals and all these things.

The very nature of any co living space means that this is a risk.

If you are thinking about politics, you are not thinking about more interesting and useful problems.

But god, politics are so fucking fun. Perhaps it's my own perversion, or perhaps it's a product of us being so wired for drama and competition and violence and tribalism. Regardless, few things get me as excited as a little game of politics against another party, logical warfare, and absolutely smashing their position.

Fun, but I can't imagine it's very good or productive. In a moral system with a high value on human flourishing and scientific progress, I'd think politics would be a necessary evil at best.

No solution yet, just an observation. Perhaps a powerful executive function with a baked in democratic release valve in case the executive really did go off the rails might work at small scales.

[Discipline as a Cookbook]

So, where's all this going?

I'm really just wondering what a sort of modern monastery would look like, if it's even needed. It seems that the internet is a particularly challenging modern opponent to focus and working on interesting problems, but is also so powerful when wielded correctly.

The tangent on politics happened, well, because politics can get in the way of that work, too.

All of the other reflections on where I'm performing my best work, because do you even need a monastery? Or is the monastery of one's mind enough?

In the past, Monasteries got us through Dark Ages.

I finally saw that computer I've written about, the Antikythera Mechanism.

How can something so diminished be so important?

We couldn't make this piece of hardware any where in the world for over 1500 years after the technology was lost.

What would the world be like if the secrets of antiquity weren't preserved through the dark ages in Monasteries? What other items do we have now that would have taken another thousand years if this or that manuscript wasn't written down? And if people hadn’t worked tirelessly to preserve, and then improve the art?

I know what great expense my own ability to focus in different conditions has come at. I understand why Disciplines that systematize that process would exist; not to make it any easier to unlock focus, but to make it more systematic, repeatable, scalable, and persistent.

All that said, I don't necessarily think the West is falling tomorrow; neither did the Romans, perhaps even after the Visigoths sacked the city proper.

It doesn't have to be about that, or even that grand, though. I'm just aware that there are environments and ways of thought significantly more conducive to solving important, challenging problems. I don't think that SF, with it's monoculture and bubble inducing behavior, is inherently that, nor do I think academia, bereft with politics and tenures and personal agendas, is that either.

I think each of us can cultivate our inner mental monasteries each day.

I just wonder how it becomes more repeatable and scalable and sustainable?

Some hacker houses already have an outsized impact on the world. I can take no credit for it, but C House has had pretty impressive results. Much more impressive, Nick has done amazing things with the Residency at some scale, and I don’t know if I’ve seen anything with such an incredible set of outcomes as HF0.

I wonder what system could spawn out of these things that could take the best of the modern world while lasting for decades, centuries, or millennia, as monasteries have?

Live Deeply,

If you liked this post, please subscribe - I’m here each week, even when galavanting in Europe.

2026.08.02

CLXIII

[Perfectionism With Haste; History Exams and Poetry; Every Single Sunday; What Writing taught me about B2B SaaS; Finding the Right Constraints; Faster Horses]

Thesis: Perfectionism without constraints becomes procrastination and is counter productive.

[Perfectionism With Haste]

For the past 3 or 4 months, I have been obsessing about WhiteWhale's product with a level of focus and attention to detail that I haven’t before.

Part of this is perhaps because we altered the structure of our business at the start of March to let me focus more on product and Jack more on sales and marketing. Not an absolute separation of concerns, but one that has been quite helpful.

The perfectionist in me has once again come out a lot more as I'm prioritizing the product and customer experience above all else.

Still, it can’t all be perfectionism, or you’d never ship anything. You have to move, and constraints help you do that. There's this strange duality that exists between two conflicting ideas:

• Perfection: everything should be of the upmost quality possible. Inspiration: Good enough never is.*

• Haste: Waiting til something is perfect to share it with the world is massively dangerous. Inspiration: Hesitation always gets you killed.**

In reality, you have to mix an attitude of perfectionism and haste together. Otherwise, your perfectionism can become procrastination. Maybe something more practical is:

• Perfection With Haste: Deliver the best possible product given constraints. Haste, because time if often the constraint.

_{*My father repeats this one a lot; not sure the absolute source but some people think it’s Debbi Fields}

_{**Didn’t cite - J Cole saying this in Sunset is why I’m thinking of this but it’s too generic for me to credit him with it.}

[History Exams and Poetry]

Innately, I'm a perfectionist.

In 7th grade, my history teacher had to have an intervention with me.

Our exams were answered with freeform text, but had to be complete thoughts and sentences. One word answers wouldn't suffice.

Brevity wasn't my problem, though; I went the other way, scrawling text all over the page in the smallest of scripts, literally wrapping it in the margins sideways to fit more. I didn't want to miss a single piece of context. Turns out, my teacher didn't want to read all of that. So, she sat me down and told me my answers could be far more brief and I’d be fine.

Later, when I was 18, I fancied myself a poet and became one; my perfectionism and obsession came out in full force.

Coming up on high school graduation, the spring and summer left me as an obsessive, somewhat neurotic and certainly restless person with no objective function: no homework, no essays, no projects, etc.

Now, don't get me wrong, I was living life and keeping busy: I ended up with my first girlfriend then, had a part time job, worked out, was going to quite a few parties, and was trading, too.

Still, my mind seems to require some big, challenging problem to grapple with at all times. Somewhere in there, I spontaneously decided I'd publish a poetry book.

I began obsessing about it. I would jot down thoughts and ideas while at a stop light driving. I kept a note pad by my bed and would wake up to note something down the minute I had something.

I'd order the poems, re order them, play with syntax, capitalization. I paid a friend to make a cover for me. Ordered a copy, held it in my hands and edited it inline.

I even had planned out a whole trilogy; I published the second instance, but never completed the third.

I wish I knew how much time I actually spent on it, but it was undoubtedly hundreds of hours.

At some point, I launched it. I need to re read it, but I do think it was quite good, and polished, even if amazingly vulgar (I was, and am, a Bukowski fanboy). Some of what I was going for was too subtle for anyone but a handful of people to see, which means it was too subtle, but I don't look back and think of it as low quality.

Interestingly, when I published it, I thought people would just buy because of the quality; this could not be further from the truth. The 80 or so copies I sold in the first 3 months were largely via hustle in the dorms and at parties.

The point is that the obsession was there, but from a commercial perspective, the results weren’t.

In other words, perfectionism is not enough. If I wanted the book to be read by more people, another round of edits likely would have had no more effect than the last one did.

Obsession: practicing shots weighted with kettlebells on an inappropriate surface in a gym until there were holes in my socks, oops (did not take as long as you might think)

[Every Single Sunday]

The perfectionism of my earlier days sometimes created good work, but I think this blog is a much better example of mixing perfectionism with haste.

Every Sunday morning, I wake up, and start writing.

For 3 or 4 hours, I'll guzzle coffee and write until it is done. I allow for very limited distractions and no phone usage, although I will admit that occasionally the conversation at the table next to mine in the coffee shop piques my attention.

I'll perhaps take one or two breaks, sometimes more. One time recently, I realized that I had sat down for 3 hours without standing up until it was done. Incidentally, this means I failed to drink enough water.

For those 3 or 4 hours, I am obsessed about and laser focused on the blog: writing, editing, deleting, rewriting, re ordering, tweaking. It's a non stop flurry of action and reflection.

At some point, though, it's done, and I publish it.

If I re read it later, I'll cringe at typos or the occasional (or frequent) run on sentence. And oh, how do I abhor unflagged inconsistencies.

In the 3 or 4 hours I'm going at it, I'm locked in, chasing perfection. Then, when it’s done, it’s done.

If I really wanted 'perfection', or at the least, something closer to the best possible output I could create on a given subject in a given number of words, I'm sure it would take 2 or 3 or maybe even 4 or 5 times as long as this does.

But, that's not what I'm after - I want the best possible output given the constraint of 3 or 4 or sometimes 5 hours of input.

Perfection within constraints; perfection with haste.

[What Writing taught me about B2B SaaS]

Over the last maybe 3 or 4 months now, I have been absolutely obsessing about WhiteWhale's product.

It blows my mind the sheer magnitude of things that can impede someone's enjoyment and pleasure when using a software product. Small things, like if you put inputs into some pop up menu, then click off without saving, they'll be confused as to why the thing didn't save.

A defensive person might say, well, that’s their fault, they should push the save button. If you’re building software for a lot of people to use, though, you kind of have to assume that if it’s not insanely obvious, some percentage of your users will always mess up on it.

Which is easy enough to solve for; depending on the inputs, maybe you either auto save the inputs or show a warning when they click off saying it won't save unless they click a button.

But, not solving it adds up... it will give the impression of immaturity and/or bugginess, and users really can have a death by a thousand small cuts!

Then, there are the bigger things, too: how long does it take the user to get their first bit of value? How much effort do they have to put in to KEEP getting value? If it's working in the background, are you explicitly communicating what it's doing so they can perceive the value more easily?

These things, like the small thing above, are all solvable, but they take time, and are often the product of many smaller changes to get there.

As an example, if that above pop up menu is a prerequisite to getting value, then having them be able to click off without inputs saving can actually be a very big problem indeed! And, if there are a lot of steps to get value, then their could be a number of similar problems hidden along the way.

That said, if those inputs are needed to get value, why are they needed? Can you make some of them default to settings good enough for 80% of users? Can you set defaults based on other inputs the user already gave you? Best, can you delete it altogether?

With so many different possible paths to take and no realistic way to have the time to explore all of them, I actually think there are probably a few quite different 'perfect' versions of any product, perfect in their own right and under certain constraints.

The issue is, if WhiteWhale would've been hidden in stealth for the last 2 years and only had 3 customers until it was 'perfect', no matter how 'perfect' we tried to make it, we'd just be scrawling endlessly in the margin of the exam to answer some question that maybe actually didn't matter at all!

What if those 3 customers weren't actually like any future ones? What if they each had some organizational flaw that we spent time building into the product?

Any notion of perfection changes with the constraints you put on the product, and the constraints change based on how you change the business and who you target and who you won't even let buy the product.

[Finding the Right Constraints]

The business you’re building around the product necessitates some of the constraints. I do think product is subservient to the business’s ability to grow and sustain itself.

If it weren't, then we'd all build in vacuums forever until our perfect little toy was done. And since you couldn't see users using it, it probably wouldn't be so useful!

You need a product that supports a growing business (assuming you want to build a business that grows, which most people do!). Perhaps at some point it's sufficient to optimize the product for cash flow and let the business sustain itself in perpetuity, but I'm not there yet, so that's not what I'm writing about.

And since the amount of time you have to build is always limited, time is always a constraint. Hence haste.

Perhaps the best question is, will obsessing about this product or feature or bug or issue let me grow the business more? By the way, as anyone who has hit a churn wall will tell you, not losing customers is part of growing a business...

Remember, you're building leverage.

[Faster Horses]

I also believe you need to have the Steve Jobs hubris, the approach of saying the customer doesn't know what they want.

We return to that oft repeated Henry Ford phrase:

I love my customers, and wouldn't be here without them. But really, they don't always know what they want, let alone need. Perhaps it sounds arrogant to you, but it’s Jack & I’s job to understand what our customers actually need and give it to them in the simplest possible way.

That’s the trade being made; you give us money, we solve your problems, sometimes even ones you don’t know you have!

When someone has a problem, sometimes they might communicate a problem one or two layers up from the root issue. 'This part of the ui is confusing' could mean that part of the ui is confusing, or it could mean the fact that the page exists entirely is confusing. Those are two very different things with very different solutions!

So, I think it's actually necessary to have some of the swagger of seeing through customer requests to the underlying issues. This is how you build what actually matters, not what looks like it might.

This is where you direct your obsessive, perfectionist energies.

I’m here every week talking about SaaS & writing & philosophy & the West & all that fun stuff. If you enjoyed this post, please subscribe for more.

[Empathy]

Perfection is necessary, but not sufficient, to build something that gives value to people.

It is possible to pour obsessive effort and energy ceaselessly into one thing without anybody consuming it, but that's a hobby, or an art, if you want. More perfectionism with my poetry book would not have necessarily made more people read it.

Perhaps that’s how starving artists are born.

Either way, perfection is not all that is needed to start a business.

There are endless things you can change, but you are trying to grow, you need to keep focusing your obsessive energy and efforts on the highest leverage one. Right now for WhiteWhale, that means an absolute obsession with product… guided by constantly asking what will give more value to the customer and make them stay longer?

As I've said before, a really great product truly is a moat.

Live Deeply,

2026.07.26

CLXII

[Happy Birthday; Fleeing Como; 4 Months of no Revenue; The First Customers; The Danger of Listening; Creeping to $100K; Getting Opinionated; Consulting Ceiling; A Product that Sells Itself; My Way]

Thesis: Starting a company is hard & non linear. I still don’t know what I’m doing, but you might find something from my journey useful.

[Happy Birthday]

My startup, WhiteWhale, turned 2 years old last week.

In that time, we've gone from no product, no brand, no customers, and a $3,000 investment from ourselves into a profitable, bootstrapped business that pays the bills & supports over 700 users.

Admittedly, just about everything has taken longer than either myself or my cofounder, Jack, thought it would. Likewise, we're still far from the ultimate, private destination we have for ourselves and the business.

But, we've come a long way, and for almost a year, have had a business that could support us.

This is a good moment to retell parts of the story in something of a consistent narrative so you can get a sense for the shape of the journey, and glimpse both a few moments that made a big difference as well as long, boring, arduous lulls in between.

That said, if you want to start a business, I actually don't know if you should copy me or not. I’m sure some of the things we’ve done should be repeated, some should be avoided, and for even more, the jury’s still out.

Still, I’ll make an attempt to tie cause and effect together without being fooled by randomness; ultimately, you must draw your own conclusions.

At best, you’ll find a useful lessons in here. At worst, it should be entertaining.

[Fleeing Como]

What was it like to start a business?

In the beginning, there was guilt.

Two days after we were able to start the business, I had a one way ticket to Europe to see the Olympics in Paris & vagabound.

Even before I left, I was conflicted: why was I going when there was so much work to do that, for legal reasons, I was unable to do for the last couple of months prior?

Still, I went off to Europe. I saw a couple of events at the Olympics, but found myself shipping code during a boxing match.

Maybe week later, I was sitting in an apartment in Como—I spent one day there coding for almost 12 hours.

We weren't making money, but I was spending money to be there.

Why? I could code for 12 hours from anywhere… and, some such places would be free.

I had a conversation with another founder on the phone, and he asked me the same question I was asking myself: "Why are you there? It won't increase your odds of survival."

So, I booked a flight out of Milan for a couple days later. Before I left, I enjoyed a swim in that brisk mountain water and a few last meals in the winding alleys, had a farewell tryst, and then flew home to live with my parent’s in Michigan.

Still is one of the most beautiful places I’ve been.

[4 Months of no Revenue]

In the start, most of my time was spent coding or talking to prospects & beta users who we wanted to one day have as customers.

Lots of caffeine, not enough sleep.

Really, not much has changed, although now, the conversation are much more often with the people who are actually paying us.

My breaks were usually me pacing outside in my back yard, reading a book. I got into Asimov at the time, the Foundation books. Funnily enough, it impacted my engineering: the notion of 'magic' being the product of extreme and aggressively applied efficiency fascinated me.

I had a lot of hubris. The way I was engineering the product was not the way any one else I spoke with thought I should. It was a lot of "Why are you doing x yourself, you should pay for an API like y".

I ignored them: my hubris proved critical. A lot of our powerful features, like daily alerts, look a like accounts, and even our comparatively superior data quality, eventually came from the arrogance that told them no and myself yes.

Perhaps, de facto, the difference between unfounded and founded hubris is if you end up being right.

This was a period of intense focus, and much trial and error. On top of that, there was a lot of confusion and iteration & frustration. They all stemmed from expectations that were misaligned with reality.

We started in July, 2024. We thought we'd have $100K ARR by the end of the year. We didn't even get our first customer to pay us $100 until November.

[The First Customers]

At the time we started, Jack really began experimenting with LinkedIn content to see what would get attention.

In November, he had his first post go viral. It resulted in something like 100 people booking calls with us and me having 70 demos in 8 days.

Even after all of these calls, and dozens more the months before, we ultimately only closed 3 people for $400 MRR total, all of which took one or more follow up calls. The first came through at the end of November, 4 months after we started.

The way this timed out, I took over a hundred sales calls before we closed our first customer.

That's a lot of work with no reward. All of that is with over a hundred discovery calls in the sales space earlier in the calendar year.

I'm emphasizing this so much, because if you are starting a company and you talk to 10 or even 20 potential customers and nobody hands you their credit card, tough shit. Either I’m exceptionally incompetent, or you’re severely underestimating how much volume you need to do.

There are a lot of things that were holding us back, of course: I was bad at sales, I didn't have confidence in our product, people taking calls with us were often times just curious to see the new circus in town.

All of that starts to go away, with time and focused effort.

[The Danger of Listening]

Once we had our first few customers, we started getting a lot more information about what people thought they wanted us to build.

This is both a blessing and a curse. A blessing, in that you're no longer building in a vacuum, and you can see if people actually want the thing you have. A curse, in that a lot of people will want different things, and if you listen to all of them all the time, you'll end up with a Frankenstein of a product that's not so good at a lot of things rather than a product that is really, really good at one thing.

As we started listening to the people paying us more than the people who weren't paying us, the product began taking shape.

It became clear that we needed a lot more data than just what was found on a company's website, so we started adding in some social posts, news feeds, and SEC Filings. Rather than pay for APIs for all of this stuff, I wrote code at the end of 2024 to scrape a lot of this ourselves that is still running today with few updates and tweaks.

On top of that, Jack was iterating on our front end and we were trying out features like weekly and daily alerts, along with even finding people at the companies our clients were targeting.

Some of this stuck, some of this didn't. This is why more time with customers is a double edged sword, a necessary evil: both potential users and paying customers will tell you that you need certain features. Just because they say that and they give you a positive reaction when you show it to them, it doesn't mean they'll actually use it or that it's good for your business.

I think this is especially true very early; in January of 2025, I think we had a couple thousand dollars in revenue, maybe three. That's not a whole lot of people paying us a lot of money, but it’s far more attention and opinions than we’d had before.

[Creeping to $100K]

The first half of 2025 was a lot of iterating on both the front end and the back end. Going from MVP code that worked to something that would be more scaleable and wouldn't break all the time.

This is crazy to say, but in January 2025, I was still running prod as a bash script that opened tmux windows before I learned the joys of systemd & eventually ansible. I was beginning to realize how much I didn’t know about engineering and how much there was to learn and improve.

Jack had another viral post or two that resulted in more call marathons; win rates were still low, but they were improving.

We thought we'd hit $100K ARR in January; we didn't hit it til June or July, literally a year in.

Like closing the first customers, we were happy than we got here, but I wouldn't classify it as insane euphoria. Personally, it had this weird dichotomy of both coming later than I thought while also feeling like it was always inevitable when it finally did hit.*

_{*This is a strange contradiction, but it does characterize founders a lot.}

[Getting Opinionated]

Over the summer 2025, we made some decisions in rapid succession that together made our product & offering a lot more attractive:

• We doubled the unit price & raised the minimum

• We cut some tangential features, like any people data we had, doubling down on our core value prop of company level signals

• We got a lot more assertive & convicted on sales calls

• We qualified harder in our meeting booking, making people acknowledge price in advance

These decisions are an important piece of what we did, because a lot of it is so counter intuitive. We cut a feature everybody said they wanted (people data), we raised prices, and made it harder to talk to us.

I think what happened is that we came across as a lot more mature than we did before; it’s hard, when you’re just starting, to confidently say no to anything at all. But, when you do it, people respect you more & take you more seriously.

Ultimately, this triggered a big growth spike coming off the summer into the Fall. October in particular was an amazing and exhilarating month.

Things were breaking, though; we were spending tons of time on sales calls and customer support and success. A champagne problem, truly, but a lot to deal with.

It was at the end of the summer that we started paying ourselves, and in the fall that I moved out of my parent's house to split a room in a hacker house.

[Consulting Ceiling]

Our constrained bandwidth revealed a product weakness - we often needed to give heavy support & coaching to the clients for them to get success with our product. To deal with this, we basically had two choices:

• Raise prices & commitment & get paid more for our time

• Keep prices low / lower them & automate more of the inputs we were doing

At first, we went with the former: we started selling quarterly and annual contracts with 4 figure monthly minimums.

At the time, we thought this was the best solve. When our product works, it really works. But to get it to work, there was a lot of configuration and tweaking and fine tuning, and then the customers actually had to inject it into a process.

We thought that by charging more and getting commitments, we'd have more time to help the customers really get success with the product.

I think this class of solve would have worked if we charged even more and formalized the process further: eg, for a quarterly pilot, make it $10K, not $3K, and require weekly calls with the decision maker & trainings and all that fun stuff.

If we did this, though, then the business would've been bound by man power. To add more customers, we'd need more sales headcount. We'd basically be giving ourselves a job or tying growth to hiring!

Regardless, we found ourselves in this weird middle channel where implementation required effort but the cost wasn't significant enough to make the counter party really commit to it; it was a lot of effort on our end that resulted in aborted deals, and we weren’t able to do enough volume.

We were making more money than we ever had, but we could see that we had given ourselves a ceiling.

[A Product that Sells Itself]

So, as I haven't shut up about since April, we decided to go the other way: lower minimum price, lower commitment, with a product that just works.

That last part is the hard part: a product that just works. When you’re doing product led growth, I don’t think it's enough to have something that in some conditions gets your client ROI with certain configurations; we're learning that you really, really need to make it fool proof and incredibly simple to configure and use.

That's what we've been grinding away at lately: we want any customer who is a good fit & buys on our site to get value out of WhiteWhale.

Jury's still out on how much work is left between where we are and where we want to be, but we're confident we can figure it out.

I tend not to be the sort of person who carries many regrets, but with the benefit of hindsight, I'll make two observations:

• Doing product led growth forces you to be a lot more brutally honest with how easy your product is to use, because you're not always gating access to it via a sales call / training / framing.

• Once you start developing a product that people can just buy, there are a LOT of automations and features that are table stakes for success. They take time and effort to identify and build out. Even moving at break neck speed, there is a lot of ground to cover.

So, part of me does wish we would've started the product led growth journey sooner; I would've been able to spend more time aggressively improving the product and less time on CS / sales sooner.

If you enjoyed my ramblings, I’m here on about startups, software, the future, philosophy, the West, and more, every Sunday - please subscribe!

[My Way]

I tell you this story now specifically because it is strange and unorthodox.

We didn't raise capital, we didn't get into a fancy incubator, we weren't born good at sales or marketing; hell, I was a very green programmer when we started, even if I have something of a natural inclination toward it.

It also isn't exactly linear and involves a lot of long and boring parts and mistakes, some of which we've corrected, some of which I’m sure I don't even realize we're still making. And, of course, we're still deep in the trenches; there are things I haven't shared yet on purpose, and other things that I don't even realize are relevant to share yet.

It's hand to hand combat, but at least at some point you get to start paying yourself.

I'm grateful for the opportunity to be here and fight these fights, and I'm grateful for everyone, including my parents & Jack, who has made it possible.

Even though I’ve interspersed My Way through this post, I don’t at all feel like we’re near the final curtain call; rather, I can't help but feel we're a few decisions and less than a hundred hours of intense work from coming to a close on Act I of the WhiteWhale story.

I’m quite excited for Act II.

Live Deeply,

2026.07.19

CLXI

[Meditating in Front of a TV; Stuck in Flavortown; You're Buying Beaches and Women, not Coronas; Under the Influence; The Heat Around the Corner...; The Utility of Competence Porn; The Library of Competence Porn; Be Inspired]

Thesis: The Media you consume has a powerful influence on your world view; if you’re intentional, you can leverage it to be inspired.

[Meditating in Front of a TV]

When consumed, media is a super powerful tool. It can comfort, it can inspire, it can distract, it can erode.

I found myself in front of a television last week and was awe struck by it's power to consume my attention.

Try as I might, I had a terrifically hard time looking away. It was like a fly trap that snapped out and ensnared me in spite of my ostensible disinterest in the subject it was televising.

Still, Media has it's utility.

Specifically, there is a particular kind of media that inspires me: Competence Porn. When I say Competence Porn, I mean some content (typically a film, a novel, a biography, a documentary, or a series) that follows one or more competent people as they complete complex & challenging tasks with low probability of success.

The film Heat is a great example of this, and I’ll list over a dozen later on.

Media has such a profound impact on our world views. If you’re going to consume media, you might as well consume stuff that you believe will have a positive impact on how you see the world & the outcome you’re after.

[Stuck in Flavortown]

Last week, I ended up in a room with a TV on at all times.

The conversation was lower frequency, and I was resolved to be reflecting while I wasn’t talking, rather than reading or coding or one of my usual outlets.

What I noticed happening, though, was how natural it was for my attention to be drawn toward the TV, even when I didn't want it to - even when I was actively trying not to focus on it.

Outside of social media, television programming has to be one of the most well optimized things to get anyone's attention.

Against my better judgement, I found myself invested in the trials of Chicago man vs a Kentucky woman as they cooked off in Guy Fieri's Flavortown.

From a high level, it’s pretty incredible that I would care about this programming just because it was on. But really, the show had a mix of things that might appeal to nearly everyone:

• Every contestant always has an emotional, heartwarming story

• The food looked delectable

• Guy Fieri is funny enough

• We all have a latent desire for recognition and reward that’s channeled through identifying with the contestants

• Of course the competition and drama and suspense keep us on the hook

As well see, I'm not inherently anti television, even if I think it's over reached; I'm in awe at the perfection of the medium as an attention trap.

I would be thinking to myself, "I'm going to reflect on this concept" and 6 minutes later realized I was totally invested in whether or not the woman from Kentucky was going to get her crumpets on the dish in time.

[You're Buying Beaches and Women, not Coronas]

This is without mentioning those damned commercials. So short, so efficient. So perfect, really: they do exactly what they need to do.

Nearly every single one was selling the outcome of their product or service, rather than the product or service:

• You aren't buying a corona, you're buying a great time in the sun, outside, probably at the beach, with beautiful women and handsome men

• You're not buying vitamins for your kids, you're buying the right to consider yourself a thoughtful and caring parent (the ads I'm thinking of were much more targeted at this than the health of the kids)

• You're not buying RAID, you're buying a wasp free party and the recognition of your friends and family for removing those pests

• Even the political campaign ads weren't asking you to vote for them, they were saying you DESERVED no state income tax or the state NEEDED a strong manufacturing base (you’ll get jobs and a better economy), and if you vote for me, you’ll get it!

This is incredible marketing, probably close to 'perfect' for this time and place and the medium. So well optimized it's shocking.

[Under the Influence]

All that said, not only is it near impossible to yank your attention away from a lot of modern Media, but when you do watch it, you’re sometimes subtly, often times overtly, effecting your world view.

If you consume what is marketed, then as a default, unless you’re actively guarding against it, you will be subtly influenced by the messaging whoever is doing the marketing wants you to believe.

If you consume a bunch of contents where the people with the money are always the villains, you’ll be pressured to believe that money is bad.

Loosely, I think a powerful world view is one that actually reflects cause and effect in the world. Meaning, if you do x or y or z, then a or b or c will happen.

If the media you consume illustrates such a powerful worldview, you’ll be more inspired to actually copy that world view and draw on it when you need to.

Of course, this world view should be ethically aligned, too, but we’re taking that as a prior, for now; in reality, it’s a concept that warrants it’s own post.

[The Heat Around the Corner...]

Last night, we watched part of Heat.

If you haven't seen it, Heat is, in my view, one of the 10 greatest films of all time. Both Al Pacino's character (a detective) and Robert De Niro's character (a crime boss) are not just incredibly well played, but are both insanely competent and capable and perfect foils* for one another.

Without spoiling anything, the film is a game of cat and mouse with each man perpetually out maneuvering the other and really, the world.

Okay, I will spoil the first 20 minutes, so don't read the next two text block if you care that much:

One of the first scenes is an almost perfectly executed heist on an armored car by De Niro's character, even perfectly predicting the LAPD's response time like clockwork. Then, when Al Pacino's character shows up on the scene to investigate, he's almost immediately able to see every single relevant detail that would give him insight into the crime and it's perpetrators, and wastes 0 time assigning someone to every single possible thread to collect information.

This is Competence Porn, one of the most enjoyable and useful sorts of films in my view.

_{*I've always thought which party was the foil was a matter of preference}

[The Utility of Competence Porn]

Surely, you've had a situation in which it feels like everything is going tits up.*

When that happens, whether it be a prod fire, or a customer getting angry at you, or receiving your 4th cease and desist letter, or just a generally overwhelming amount of small things all going wrong at once, I really think you have two primary options to respond:

1. Getting nervous, panicking, over reacting, shutting down

2. Taking a deep breath, calming down, assessing the situation, and proceeding with the actions that maximize your odds of winning

Competence porn can be helpful to push you toward the latter. You remember what some character or historical figure you really like would do, and then you do it.

This might sound incredibly childish and fantastical, but guess what? I don't care!

I don't think you should care, either.

When I was like 12 I would always ask myself "What would Paul Muad'Dib do?" if I was having a hard time with other kids or something. And guess what, it helps!

Now, if I wrote some code that's hitting a scaling constraint or a nasty edge case that needs to be immediately addressed, I ask myself what would Pham Nuwen do? Obviously cooly and calmly code his way out of it.

It's not cringe to be inspired. As a matter of fact, it's cringe NOT to be inspired.

Even now, as I was writing out those 2 options above, I was thinking about a Jocko Willink anecdote. I can't remember if it was from one of his books or an interview. Basically, he said that when deployed, the Marines were calm when nothing was happening and the Seals ran around like they had sticks up their ass checking everything, restlessly preparing. Then, when conflict did alive, the Marines would be more inclined to lose their cool while the Seals became incredibly calm: after all, the uncertainty was gone. Threat was identified. It's go time.

Am I a navy seal? HELL NO. I'm a dude who ships code and sells SaaS on the internet and lives in a hacker house.

But if given the choice, why would not want to be more like the seals?

_{*Had to link this because I found out a week or two ago that this was not exactly a very common phrase outside of certain circles}

[The Library of Competence Porn]

As promised, here is a library of some of my favorite Competence Porn:

• Heat: Both characters are insanely competent

• First 3 Foundation Books: Thousand year plans with nested plans, so much dialogue like: "My space ships have encircled your spaceships, you must surrender" "Well, actually, my men are already ON your spaceships. Your Fleet Captain surrendered 20 minutes ago!"

• James Bond: the earlier films are a bit 'goofy', but the Craig ones are pure high octane problem solving & quick thinking under pressure

• Mad Men: Don Draper's key redeeming quality is that he's competent

• Dune: A lot of the characters and factions are playing generational chess

• A Deepness in the Sky: thousand year long schemes, hacking, coups, and leveraging technology to get insane edges

• Julius Caesar's Journals from the Gallic Wars: 7 years of persistence before winning. 2 or 3 moments of sheer military genius. I have a canvas print of the surrender of Vercingetorix.

• Elon Musk by Walter Isaacson: Moment after moment of intensity and focus, include changing the design of a rocket less than 24 hours before a launch & it working

• Dark Knight Trilogy: Most Nolan films are full of competence porn, but these 3 in particular are pretty awesome

• King Icahn: Schemes like the man sleeping in the middle of the day to show up late to a negotiation so he was fresh and awake as he drove them all the way to 3am deals while the counterparties were exhausted

• Mike Duncan’s Podcast on the French Revolution: A handful of the players were quite unstoppable, like Metternich & Lafyette & Talleyrand (so good that his name is literally a noun for an expert diplomat)

• The Killing by Kubrick: An insanely tight and enjoyable heist film

• WestWorld S1-3: Deteriorates over time, never watched s4, but there are some characters who are great at both planning and thinking on their feet

• Sicario: Matt Graves & Alejandro are both forces of nature. Sicario 2 is also good but I kind of feel like they leaned too hard into the best parts of the first one

• Cash McCall: A super efficient businessman does a buy out of a company partly to get with the owner's daughter

• Succession: This one is interesting because a lot of it is ironic and there is a laughable amount of cringe worthy incompetence, but some of the characters have moments of brilliance and others are outright sharks

• Conquest of the Useless: Five years of Herzog getting a 200 ton boat over a mountain in the Amazon. 'Should've failed' a number of times. Didn't.

If you consume any of the content there, let me know what you think.

I have a canvas print of this in my room at home

I’m here every week yapping about tech or media or culture or the West - subscribe to never miss a beat.

[Be Inspired]

Media is so well optimized to grab your attention that it can be quite hard to look away from.

Not all media is created equal, though. Some locks you in and drags you into a world you don't really care to be in.

Other media can actually inspire.

It might be laughable to you to seek inspiration from fiction, but it's what humanity has been doing ever since we shared the first creation stories and myths.

It's not something to be ashamed of, nor is it something to be afraid to lean into.

In my last negotiation, which took ~11 days start to finish (the counter party might say it was only 3 hours), I actually treated it like a character in an Asimov book would, and I got quite a good outcome. And not just for myself: in all honesty, I think everybody did.

Live Deeply,

2026.07.12

CLX

[No Investment Banker Dinners; Auto Pilot Acquisitions; PMF is a Gradient; A Churn Wall; Too Much Value?; NRR; Lead Bullets; Silver Bullet]

Thesis: Product market fit is a gradient that you need to keep iteratively working toward, and solving churn is a big part of it.

[No Investment Banker Dinners]

WhiteWhale is coming up on it's 2 year anniversary.

And really, the underlying problem technical problem is something I've been involved in directly since 2022, and more indirectly for a good chunk of my life.

Given all that, sometimes I think we’re a bit delusional for continuing on the quest even though we don’t have Product Market Fit (PMF) yet.

For those unfamiliar, Marc Andreessen describes PMF as:

The great news is we’re certainly NOT strictly in Andreessen’s first bucket. But while we have elements of it, we’re not fully in his second bucket, yet, either, though.

This makes sense: PMF is a gradient, not a binary thing. And you get more mixed signals than in dating. A challenge is sifting through the information and finding where you land on the scale.

If you think you’re further from PMF than you are, you might over correct and pivot and fuck up the thing that was working. If you think you’re closer than you are, you might lever up to fast and get top heavy and crumble!

The good news for WhiteWhale we're closer to PMF than every other time I've written about it. We have more positive feedback and word of mouth, we’re getting intros to decision makers in our space, we're hitting all time high revenues, we're collecting customers faster than ever, and we can acquire some customers without even having to get on a sales call!

Still, we're not there. Churn is too high and usage is not expanding enough. And, the investment bankers aren’t buying us free meals, either.

So, we're going to do what we always do - tackle our problems directly and aggressively while looking for a more elegant solution that takes us a step function closer to what we want.

[Auto Pilot Acquisitions]

First things first, I've been yapping about PLG over the last 3 months now.

The really good news is that it's still actually working! My note a few weeks ago ended up not being a false alarm. There are still a lot of hiccups and optimizations, but there are some people who are buying the product and correctly provisioning it (and to be fair, we've also rewritten the product to get it a lot closer to correctly provisioning itself!).

There's still a lot of progress to be made, like getting rid of some edge case bugs that can be quite bad, or introducing a support bot, but the infrastructure is there. That, plus now having a completely automatic affiliate program makes our lives a lot easier. There are still onboarding calls for some people and Jack is still fielding plenty of old deals or prospects from his dms, but we've seen enough to know that full self service is possible.

It's just a function of further refinement and iterations on our end.

This is absolutely lovely news, because it means we aren't inherently needed to get on sales or onboarding calls to scale the business. The product can work on it's own, and in an increasing number of cases, will.

This let’s us focus less time on linear activities like sales calls and onboarding, and more time on asymmetric investments, like product work & marketing & funnels.

[PMF is a Gradient]

You can have more or less PMF. We have astronomically more PMF now than we did 2 years ago, and more than we have at any point in our history.

The mixed signals can make it really hard to understand, though. As an example, last week, in a 48 hour period, we had:

• 3 customers churn

• One customer on an annual subscription tell us that they closed a deal directly attributable to us that covered the cost of the subscription

• Another customer notice a bug in the software but then ask us how he can help promote us out in the real world

• A text from a friend saying that someone in his co working space is raving about how awesome the product is

This is without mentioning the ups and downs from hot patching product issues on the spot, etc.

The point is, the positive stuff is really positive, but the negative stuff is quite negative, too!

And you may be saying that the churn is natural; of course customers eventually leave! But, if you're not careful, churn can be so high that it fundamentally blocks your ability to grow.

The question becomes how do you get MORE of that great feedback and much less of the churn?

The all you can eat sushi place by our hacker house has unbelievable PMF

[A Churn Wall]

You can calculate your maximum revenue based on Churn and revenue growth.

When I say Churn, I mean what percent of your old revenue you lose each month.

As a fictional example, if you have $60,000 in MRR in Month 1, and then $54,000 of that revenue stays til the next month, your churn is 10% (($60,000 - $54,000) / $60,000).

Note that churn doesn’t consider the new revenue you add in the next month. If you add $6,000 in new revenue, you’d end up at $60,000 in revenue at the end of month 2—but this is literally how much you started with! So, even after doing all of the effort of adding $6,000 in revenue, you’re running in place on a treadmill!

So, Churn & revenue added each month actually show you the ceiling of your business.

If you have 10% churn and $6,000 added in net new revenue each month, your maximum revenue is something $60,000 per month.

$60,000 per month, though, is really hard to hit with those numbers. Every month that you get closer, your rate of progress slows (at $45K / month, you lose $4.5K per month and add $1.5K to end up at $46.5K. But then, the next month at $46.5K, you lose $4.65K and only increase by $1.35K!).

The only way to overcome this ‘churn wall’ is to decrease churn, increase revenue added, or both.

This is why losing 3 customers in one day is something that seriously weighs against all that other good news I mentioned above. If we add even one new customer a day, then losing 3 at once puts you at nearly 10% churn for one month's cohort*!

And yes, total churn is calculated not just with one months cohort... but, all I'm saying is I'd be telling you our total churn if it was single digitals…

Showing a weakness like this publicly on the internet might seem crazy, but, I don't care for 3 reasons:

1. It'd actually be worse to share if I said I figured it out - competitors would be more inclined to copy me!

2. I don't think a lot of new SaaS companies are doing much better, especially the ‘ai’ powered ones

3. It's a problem we're spending all of our free working hours to solve. We will solve it.

_{*One way to help understand churn is to break users up into 'cohorts', or groups with something in common, like the month in which they signed up. Even for a rapidly changing product, these people will have a more similar experience than randomly grabbing 5 people from every month!}

[Too Much Value?]

Last week, I got on the phone with a new customer. He'd been burnt by a quasi competitor charging him 9 times what we are. This competitor kept saying he could do everything and continuously fell short; he ignored my Father's oft repeated advice "Under promise, over deliver."

Anyways, this client is quite technically apt & operates a lot of stuff out a central system he built with an AI in the middle. He said two very important things.

First: "I can't build this myself."

This is quite awesome coming from the kind of guy who is always asking what he can build and what he can't; it's good validation that the challenge of recreating our software is far more expensive than the $200 he's currently paying.

Second: "You're giving me too much value."

This is something we've always known and something my sales coach has warned against. It's the reason we started doing quarterlies and annuals at one point: if we're enriching and ranking an account list, we give a lot of value upfront!

Especially in his case; he had 36,000 records, and we showed him the best 1,400. Ranking 36,000 records would be insanely expensive for many of our competitors. Granted, we only let him see the in depth data on 300 of the accounts this month, but he still knows the names of the other 1100 & gets some useful info about them.*

Similarly, in the last 2 weeks, we heard from a rep at a client that used us for 3 months: he told us that after they got rid of us, the client still closed a deal off of our data, giving them insane ROI.

In some ways, this might be the best news we've gotten... some portion of our customers are churning because they're getting everything they need right away. Meaning not only is it valuable, but it’s too much value at once.

[NRR]

Net Revenue Retention (NRR) is another concept related to Churn; it's how much of last month's revenue you kept on a percentage basis. Meaning, if your churn is 10%, NRR is 90%.

However, NRR can be positive, too. If you have 10 customers each paying $200 per month, and you have one leave but 2 double their subscriptions, NRR becomes 110% (($2000 - $200 + $200 + $200) / $2000).

If you have positive NRR and do nothing, the business will just grow. Moreover, this means you can neutralize churn by up selling your best customers (the ones who want to brag about you at conferences) rather than just by reducing the number of customers who leave you.

Every customer who doubles thei subscription buys you back one who leaves.

I bring this up here, because if some customers think we're giving too much away at the price point, this represents a very exciting opportunity to restructure the offer and up sell them.

As of now, our product is structured horrendously poorly for up sells. So, this is something we're going to fix.

[Lead Bullets]

There are so many little wins we've been able to add in the last week or two that should help reduce churn:

• Fixing a lot of the deepest & most annoying & confusing bugs

• Automating a lot of the activation for people in the PLG flow

• Adding a welcome tour that takes them through the platform with videos of Jack yapping about how to use it

• Clearly setting the expectations around the limitations of the product during said onboarding flow

• Adding a clear help center to the settings

• Adding a couple emails that go out after you buy, including one talking about success stories

Most of this stuff falls in the camp of "obvious" fixes that we were clearly missing and not prioritizing enough.

From one point of view, doing a lot more automation of the activation should help a ton. Customers who never activate have a 100% chance of churning, which is the highest of any category; so, if you make sure every one activates, then of course you'll reduce churn!

From another angle, if you view churn as a build up to some critical mass of confusions and frustrations that causes a customer to eventually leave, this stuff adds up a lot, too. It gives them pressure release valves and reduces the probability of that critical mass being hit.

Here’s a good encapsulation of this grindy, solve the issue at all costs mindset:

Subscribe here for my weekly rant that ranges from startups to philosophy to chasing truth to living a good life and creating cool things.

[Silver Bullet]

Still, below all the lead bullets, you can't help but wonder if there is a silver bullet. Is there some step function that can solve a ton of the problem at once?

At any combination of product and offer, there is some minimum churn and maximum NRR you can achieve by optimizing and making the above grindy changes mentioned.

We think there is another combination of product and offer with a much lower possible churn and higher NRR, though. Something that will feel a lot more like Mr. Andreessen’s quote above… something that might bring the bankers out in droves.

So, while we're grinding away on the lead bullets, we're also testing out a silver bullet that involves changing the front end and the way that we price and deliver value.

The goal is to build something that makes it easy for all of our customers to get as excited as our best customers. And to sell it with an offer that makes the guys who says we're giving away too much value actually feel like a more reasonable trade while still being impossible to recreate. We're pretty thoroughly convinced that these two elements, the lack of the up sell and the churn, are actually both closely related and solvable with one big change to product and offer at the same time.

This is challenging, because as I've said before, most our closest competitors haven't seem to have discover a low churn / high NRR business model, either.

The trouble is, a lot of them used VC money to grow as big as they did without ever solving this problem. You can buy a lot of things in life, including customers. If you could buy PMF, though, VC as an asset class wouldn't be in such a weird spot.

So, in other words, if Jack & I took on money for WhiteWhale, we'd have more revenue, but we'd also be losing money and likely dealing with the same issues we are now. Just at higher stakes, because if we didn't solve it fast enough, we'd run out of cash.

With a profitable, bootstrapped business, this is not a risk. We have all the time in the world, but we're still acting like the sky is falling tomorrow.

So, even being nearly two years into a business that still doesn’t have PMF, I’m more excited than ever.

Live Deeply,

2026.07.05

CLIX

[To hell with Timing; The Twilight Zone; Fallout; The MBAs Have Replaced the Pirates; Contrarian By Nature; Happy 4th]

Thesis: Taking action you believe is right gets a better return than looking for consensus or waiting for the perfect timing.

[To hell with Timing]

I got an exceptional piece of solicited advice from my Jiu Jitsu coach a week or so ago.

He told me to stop waiting for the someone else's mistake to advance my position or escape. Rather, go ahead and leverage proper form and technique on your own terms.

I have abnormally strong defense. This means I can sit in a compromised position for longer than is natural & I rarely if ever panic. I'm stuck sitting there waiting for the other person to mess up, though.

His point was I should stop waiting. I don't always need to be sneaky and clever. Sometimes, I can just very clearly and blatantly do the basic escapes with good form even if the top person is pressing me quite hard.

The same goes for attacking; I can do less waiting on them messing up and more asserting what I want to happen until it does.

I applied this advice and had immediate amazing results. It's rare you get such effective feedback.

To hell with timing; play your own game, regardless of theirs. You need to be aware of what they're doing and take advantage of it when they mess up, but you can also more aggressively do more of what you know works, too.

As a recovering finance bro, I think about markets and timing a lot.

The market my startup, WhiteWhale, is playing in, is really, really weird right now. Too much money was raised, and now there's nominal acquisitions, aggressive pivots, and overall consolidation.

This, I believe, is downstream of a Silicon Valley that has lost the spirit of the Pirate and had it replaced with the spirit of the MBA. Like I was waiting for good timing in jiu jitsu, they’re sitting they’re waiting for market consensus and someone else to make the first move.

You need to remember that you can make moves and assert yourself and take risks. That’s part of the game.

The bigger risk, at some point, is waiting & following.

[The Twilight Zone]

The market I am playing in is fucking weird. Overall, I'd call it 'data for sales & marketing teams.' But, the lines here are really thin, because a lot of 'infrastructure' companies (email sending, workflow automations) also do data and data companies do infrastructure.

In the past few months, here are some of the highlights:

• Pocus, company that raised >=$23M, was acquired by Apollo

• Unify, company that raised ~$60M, pivoted from + $16K annual contracts to $20/mo subscription with free trial, basically copying the ui/ux and feature set of Origami, company that raised $2M

• Common Room, company that raised >$50M, was acquired by Zoom, yes, Zoom, not ZoomInfo, to improve a platform that I've only ever heard mentioned once (by someone selling it).

• Warmly, having raised >$17M, was acquired by HubSpot.

• HubSpot announced a controversial change to it's TOS that means you have to opt out of them recording and selling the data in your CRM. This + Warmly, acquisition highlights their play of continue to push into the data & enrichment market.

I don't know how much Warmly, or CommonRoom were acquired for, but I will attempt to back into each in HubSpot & Zoom 10q's when they release. Rumor's about Pocus's acquisition price are not good.

[Fallout]

I find this all so weird because it feels like we're entering the fallout zone of a space where so much more money was invested by VC than was ever warranted.

If you ignore the revenue from the CRM company’s data projects, the revenue of the big data companies in the space is maybe $2B per annum (Lemlist $53M + Apollo ‘approaching’ $200M + 6Sense $200M + ZoomInfo ~$1.25B + Clay $100M + RocketReach $40M + DemandBase $200M). A lot of these numbers are self reported, unaudited, and of questionable reliability, and I’m probably missing a few big ones, but I’d be shocked if I was missing more than another $1B lying around somewhere. And, to boot, calling some of these companies truly data companies is a stretch.

Note the vast majority of this comes from one company that sells contact data that a lot of people in the market are arguing is too expensive because it’s increasingly commoditized.

Yes, $2B is a lot of money in terms of absolute dollar value. And it is big enough to make it a very attractive pie for myself & Jack; while we only play in a sliver of the $2B market (won’t even touch contact data!), every dollar we earn puts us in a better position.

But, this is NOT a lot of money for the companies going out and raising tens of millions of dollars and some even getting valued as unicorns.

If your company gets valued at even $100M, you’d have to hit $10M ARR and sell your business at 10x Revenue to break even (VC’s don’t want to break even, they want a 10x return!).

Even if you did this, if your churn was 9% a month (not uncommon in this space!), and you're not even profitable yet, whose going to buy you for 10x revenue to actually make it break even?

Put another way, if you’re value at anywhere near 9 figures, you’re already at 1/9th the public value of the company that does more than half of all of the revenue in the space. To get your investors their 10x return, you need to be worth more than the company doing half of the fucking revenue.

You'd better have some really differentiated technology, but that's hard to do when you're pressured to build the everything tool and water down any differentiation you have. And, of course, there's the greater fool theory, but that's not something I'd bet on when exit planning.

The only thing I can think of is maybe these companies are arguing they’ll expand the SaaS market or get revenue by firing a bunch of sales people? The AI SDRs claiming they do that aren’t fairing so well (source: talk to literally any sales person or sales leader instead of the founders selling it).

Any who, there’s really not much place left to turn for a lot of these companies other than an acquisition from an incumbent, but who knows how much that is for. The Corp Dev team's I've spoken with in the space have struck me as very shrewd and likely to get a favorable price. Also probably why no one ever says what they got bought for.

The more time I spend here, the worse of a market I think this is for VCs.

You already know how I feel about AI Valuations, but this does make me wonder how bad other software markets are.

[The MBAs Have Replaced the Pirates]

I think the glut of companies with such aggressive funding in this space is a down stream symptom of a perversion in Silicon Valley: it's shifted from the gun slingers and risk taking mavericks to consensus seeking MBAs & rent seekers.

When I think of entrepreneurship and risk taking, I think of:

• Steve Jobs & Wozniak building computers in a garage

• Elon making money hosting parties that he didn't attend and then years later standing in the wreckage of Falcon 1

• DropBox making a simple demo video that took the market by storm

• Mark Millet ski bumming before eventually starting Steel Dynamics and saying his steel is more high-tech than Facebook

• A midwest billionaire I once met running around the West, bangning on people's doors to buy their mineral rights and then sell it as bundle to oil companies they predicted were headed that way

Gets me fired up every time

I don't think of:

• Someone finishing a degree at a Top 20 University

• Working at either a startup that's so big it's become big tech or a management consulting job

• Moving to the Bay because tech

• Making a slide deck that explains how the thing you haven't built yet will one day be really big

• Vaguely 'doing AI' because it's the main thing right now

• More slides on tam, sam, som, moat, differentiation for a product that doesn't exist yet

• Raising a bunch of money to spend it within a year and either raise more money or go out of business.

• Go build on the cloud with docker and kubernetes because you're supposed to build on the cloud with docker and kubernetes and hey AWS gave us $100K in free money to do it!

Listen, if you need to raise money, go and raise it. But a lot of the time it just feels like the 'track you're supposed to do,' sponsored by this self reinforcing feedback loop that even if you don't need to raise in a vacuum, because other people are raising you now need to in order to keep up!

Do what you believe is right, and do it with good form. To hell with the timing.

[Contrarian By Nature]

I really believe that all of the best founders / investors / VCs don't give a fuck what anyone else is doing. They just go for it.

Sure, they're aware of consensus, but it's far from the most interesting thing they spend their time focusing on.

One of my friends with a very real product and very real business with an incredible asymmetric opportunity is raising right now. Apparently, the largest fund in the world will only invest at this stage if you 'already have a term sheet in hand' from another investor.

That is insane. That's literally waiting for consensus.

I've written about being a contrarian before. I'm starting to wonder if being a contrarian is actually a necessary condition to get a truly out sized return. If you don't take a contrarian stance, you're by definition betting on market consensus. Where's the out sized return in that?

A good trade becomes a bad trade if everyone has already placed it.

Regardless, you're certainly not waiting to move a particular way until you're 'supposed' to move or other people say the timing is absolutely perfect. You're just moving as you see fit.

At WhiteWhale, running a lot of our own scrapers was something everyone told me was a waste of time, or at the least was very confused about, as was building on prem servers and a slew of other technical decisions.

These decisions, though, are some of the big reasons we've been profitable nearly the entire time.

If we would've listened to the market's consensus (now is the time to just go full port into cloud, use apis, don't waste your time), we'd have literally the same product everyone else does, and we'd probably be losing money.

We just did this shit because we thought it was right. And so far, it's working.

And all of this is without mentioning the number of people who told us we needed to raise. Thank god we didn't, or we might be in the weird growing fallout zone starting to consume our market.

Subscribe to hear my weekly rant about tech, investing, saas, martial arts, and lord knows what else.

[Happy 4th]

You have to remember, I have no fucking idea what I'm doing.

But what I do have is a growing confidence that I will figure it out, no matter what comes.

And I'll take advice that I think is good, but I have an allergy to any prescribed path.

Maybe that's risky, but it's intoxicating, too.

The spirit of the pirate is a lot more enriching than the spirit of the MBA. It makes you feel so much more alive and so much more real.

To hell with timing, you do you. Conveniently, this is inline with the mythology of America & our independence day.

Happy 4th, keep the spirit of the pirate alive.

Live Deeply,

CLVIII

[Product is a Moat; Time Reveals Secrets; Being Patient to be Impatient; 4.5 Terabytes of Champagne Problems; You didn't build it Wrong, you built a Prototype; Money Doesn't Speed up Pregnancy; My Competition Can't See Me; Build a Great Product]

Thesis: Great product takes time to build & resultantly is a really good moat.

[Product is a Moat]

Back when I thought I needed to raise money to start a company, the contrived wisdom that the business world conspired to make an axiomatic truth is this notion that you need some 'moat' to build a startup.

Moats were taught in business school and every single entrepreneurial program I've ever participated in.

It's this notion that you can build something into your business that makes it hard for some hypothetical competitor to copy.

What always befuddles new entrepreneurs and me to this day is the claim that having a really good product is not a moat.

Rather, 'acceptable moats' are usually one of the buzzwords business people love, like 'network effects' or 'intellectual property'.*

Since AI has ostensibly made it easier to ship prototypes that people mistake for products, the 'product is not a moat crowd' is doubling down on the claim that 'distribution is the only moat'. Basically, this means that having followers and a network is the only way to make your position defensible.

Don't get me wrong, distribution is massively important - Jack's great social presence means that it's really easy for us to get eyes on our ideas and products and website. This is huge for the program!

But, much the same way as we'll see about product, this distribution came from a process of trial and error and intense focus on Jack's part to figure out what the market generally cares about content wise. Like anything worthwhile, it takes time to build.

Perhaps not realizing this, some VC / Business type brow beat young, inspired people out of the idea that a really good product is a moat because someone told them it’s not investible or something.

Yeah, you can have a shit product and great marketing and make fast money, but you can't tell me that a company with just as good marketing and a better product wouldn't eat your lunch all day.

Build great products!

It's really, really hard to make a good product. It's an iterative process that is as much as guaranteed to have failures and moments where everything just kind of breaks.

It takes time. And it takes people using earlier iterations to get there. And, as we'll see, it takes the world falling apart and fighting through the fire of things burning and breaking to get it right.

So no, having a great product isn't 'not a moat'.

I believe a great product is one of the best moats there is.

_{*I think the success of social networks have made us over index on network effects for products where it doesn’t matter. And I've always thought that intellectual property was crock of shit in SaaS - why would I file a patent explaining how I do the thing I do when it would be very hard and expensive for me to ever know if you actually copied me? Trade secrets are self evident - of course I won't share my code with my competitors! But if you talk about trade secrets, this VC / Business type gets uncomfortable and typically says that’s not really intellectual property, perhaps because they realize that it collapses into basically the same thing as saying product is the moat, but wait, they're supposed to believe that product can’t be a moat, so I guess trade secrets can’t either.}

[Time Reveals Secrets]

There are some secrets that only time reveals.

No amount of VC capital, LLMs prompting each other in loops, or yapping at networking events will show you the answer to the questions that you should be asking if you're starting a business.

Important questions, like "What features would my customers actually use?" can really only be answered by putting the feature in front of your customers and seeing if they use it. Even then, it's imperfect information - did they use it because you brought it up in conversation? Did only the customers who will churn in one month use it? Do the customers who use it actually end up spending more and staying longer?

The answers to these questions are the ones that you put back into your product.

As an example, so many WhiteWhale customers have asked us for the ability to have different lists with different signals for different use cases. We built and maintained this feature for a while, and people actually used it a lot!

But a lot of people used it TOO much. They would spend time tweaking and playing with it and over engineering with it. And the people who put the most time into this would often get the least value & churn quite quickly. They'd be caught in configuration hell rather than taking action.

The users not playing with this feature a lot generally stayed longer; unsurprisingly, getting rid of the feature increased retention.

So, in this case, neither listening to what customers wanted or even watching the short term behavior of what they used would have shown us the truth - we needed time to observe long term behavior.

I'm sure you can get an intuition about these things, but only by actually testing it. And I think that even someone who has built one great product needs a lot of time with the customer of a second product to build a great product for them, too.

For all I know, if 80% of our users were rev ops people, then maybe that feature we cut would be what actually kept them around longer!

[Being Patient to be Impatient]

In this way, building a great product is a function of staying in the game long enough to actually know what your customers care about.

The longer you're around, the more information you collect, and the better you can make your product. At some point, you've collected enough information that on some axis your product is materially better than the alternatives on the market.

And then you sprint.

For Jack & I, the 'later' in the above quote is now.

And it will be until it's not. We'll hit a plateau and then, if we keep listening, we'll have a chance to learn more than we ever could have before.

And we'll slow down to speed up again.

[4.5 Terabytes of Champagne Problems]

Asides from thinking in terms of user facing features, there are engineering decisions to make it all possible that only become clear after things break.

When we started WhiteWhale, I didn't understand what a database was. 2 years later, I manage a pretty sizable one, at 4.5 TB; more than half of that is one table!

The challenges that this creates befuddle me every 5 months or so during a growth spurt.

Over the last calendar month, we've been adding on average a customer of day, and in the last trailing quarter, added over twice that total.

This not only increased the size of the database, but more interestingly & problematically increased the frequency that it is read & written to.

This can get overwhelming engineering wise, especially when the problems appear all at once. It can feel like the world is falling apart, and you want to scream for mercy.

Until you remember that this is exactly what you signed up for: everything is going according to plan!

Oh no, you have too many customers now!

What a blessed problem to have - it really is what one of my mentors always refers to as a Champagne Problem.

So, you buckle up, switch from a reactive state to a proactive one and aggressively & quickly dispatch with the problems.

[You didn't build it Wrong, you built a Prototype]

When I first started partaking in this pattern of increasingly load leading to too much pressure on the system, I wondered what you might be wondering: did I design something earlier incorrectly? If I would have built it different from the start, would it have been more robust by the time we got to this current load?

To some extent, likely yes, but the same mentor who speaks of champagne problems also re assured me that this is somewhat of a fools errand, attempting to engineer the first draft so well that it never breaks under any conditions.

I was skeptical at first, but now having been through a number of these cycles, I'm utterly convinced he's correct.

The most basic reason cited is that by not fixating on hypothetical problems like having too many users in the future, you’re more effectively able to build a product that solves today’s problems. This is especially true when the hypothetical problem in the future is one of these champagne problems: if you have positive gross margins (gasp) and you get more customers, then you also have more money to solve those problems with!

And money often times does make the problems easier to solve. At the very least, it justifies spending the engineering time to solve them.

Another really important reason to not fret about building something perfectly for all future states is that you just don't have enough information yet. At the start, you've undoubtedly made bad assumptions. The longer you're around, the more information you collect, and the better your assumptions should become. The better your assumptions are, the more confidence you can and should have building on top of them!

As an example, early on, we thought we'd need to collect and store information on decision makers at potential target companies. As it turned out, this assumption was wrong!

Imagine if I would have spent another 80 hours making those tables and python scripts and services more robust and scalable. Maybe there would have been some system wide benefits, but as it is, the code I wrote did exactly what I needed it to: just enough for us to know if we needed it to do more!

[Money Doesn't Speed up Pregnancy]

All of this is to say that building a really powerful product that works at scale takes time and effort and is not some linear, straightforward thing.

It's an iterative process, and I'm really not so convinced that having funding or anything would have sped it up or would speed it up in the case of WhiteWhale.

These things take time and effort and trial and error and all that. Maybe with money you could speed up each step in the function, but you'd still have to be taking the steps.

I don't care how forests you burn by having LLMs spam each other in a loop, building something great takes time and the information you learn by building something bad and then something not so bad and then something okay and then something good.

I'm sure there are experts and wizards who can do it faster than Jack & myself, but even for them, time is still an input.

If a great product is a function of time input, then it takes time for a competitor to come in and copy you in a meaningful way; without time, surely they'll miss many of the important details.

And if a great product puts a buffer of time between you and the competition, well, then, that's just as much of a moat as anything.

[My Competition Can't See Me]

I've talked a lot about what I'm doing in a post that's supposed to somewhat be about competition. After all, moats imply you're keeping someone out.

This is perhaps arrogant to say, but I still don't think I have any real competition. Even the copy cat from ~5 months ago seems to have switched up a bit already (that didn't take long!).

A quasi competitor in the space that raised $60M and sold five figure annual contracts pivoted to copy another quasi competitor that raised $15M; now they both do a free trial + $20/mo base subscriptions for a chat like product with built in data apis & some sequencing tools.

Testing both, I can't imagine that either is profitable with the amount of LLM compute that goes into their free tier. And while our product doesn't do half the things either of theirs does, where it does overlap, I believe our results are much better - based on where their engineering and marketing appears to be going, they still don’t think the part we’re focused on, the research, is that important or differentiable!

So, we'll have people seriously copy us eventually and persist, but they're all fighting different battles right now.

And notice two other facts:

• We've been around for 2 years

• Our public goal of hitting $1M ARR with 2 people implies we haven't hit $1M arr yet!

Judging with VC standards, we're a fucking failure.

By the time anyone realizes they're wrong, it'll be too late, if it isn't already.

[Build a Great Product]

I don't really know why or how it became consensus that a really good product is a moat.

Maybe it has something to do with the fact that building a really good product is just really hard and takes time. Maybe people are so fixated on a silver bullet and capturing the wisdom of really successful people with some repeatable formula.

I don't know what it is. But it doesn't matter. It’s wrong, and that’s a hill I’ll die on.

It's okay to say that your moat is that your product doesn't suck.

Try to build a great product. See how long it takes for the product to actually be great - I don’t even think my product is ‘great’ yet, and it’s been two years!

Try, though. And keep trying. And see how long it takes for someone to copy you.

Live Deeply,

CLVII

[Peter Thiel in Malibu; Creating Value; Value Creation != Value Capture; Value Destruction; Don't Invest in my Business; Old AI on Hidden Data; Fighting Surveillance States; Sci-Fi Maxxing; Read SciFi]

Thesis: The best place for venture capital is investing in value creating technologies that need capital to change the world.

[Peter Thiel in Malibu]

I've ranted a lot about how I think that while LLMs are a cool and sometimes useful tool, but that the financial investment into them is completely overblown and the negative impact of the way people use LLMs likely on the whole outweighs the positive effects.

Last week, I got an interesting question last week about where I think the money and investment should go instead.

It was actually very similar to the question that I personally asked Peter Thiel that one time I got to meet him at a gala in Malibu in Fall 2021.*

He once said something about over investing in the 'bit' rather than the 'atom.’ Meaning, we're overspending on digital technology rather than 'real world' physical stuff.

In a very convoluted and long winded way, I asked him if he still thought that was true in 2021, and he said something along the lines of 'kind of yes, but bit vs atom is also not the only paradigm that matters.'

DE** asked me basically the same question: I keep ranting about AI and LLMs wasting money, but still seem to be a proponent of investment in digital technology. Why?

I bring up the Malibu story because it occurred to me that my answer to DE is maybe actually what Thiel meant when he answered me 5 years ago. But also, maybe not.

I think digital technology actually enables pretty amazing things in terms of physical hardware & unlocks functionality and gains otherwise impossible with strictly hardware devices.

So we’ll talk about that a bit today with 2 real examples, but more broadly, I’ll explain from a high level some of the places I think it’d be interesting to deploy venture dollars.

We cover:

• Value Creation, Value Capture, & Value Destruction

• Why you'd never want to invest in my startup, WhiteWhale

• 2 examples of digital tech enabling physical hardware to create genuine value

• A list of other really cool technologies, like space ships and nuclear energy.

Before we begin, remember: I'm just a guy building cool shit & sharing strong opinions.

_{*Funnily enough, a couple days after that, BH & I ended up a hacker house in SF full of OpenAI engineers. Probably some irony given my questions & current views; at the time, I had no idea what they were doing.}

_{**If you haven’t inferred yet, I use personal friend’s initials when I talk about them. If you know who it is, you know who it is, I guess.}

[Creating Value]

First and foremost, I really believe that anything you invest in should actually be creating value. I'm actually pleased that this week, a friend commented that I'm one of the few people she knows who is both very focused on wealth accumulation while also being focused on value creation.

Value creation is this idea that the product or service you sell, when used, makes more of something that is worth something to someone, and ideally something that makes the world more resource rich and enjoyable.

With my own simple startup, WhiteWhale, as example, we started with no code, no product etc. We traded our time and effort to make a piece of software that wasn't there before. Now, people can use our software to 'get value.'

If a sales or marketing person is able to use our software to, say, create a $60,000 deal that they have have a 33% chance of winning (so the Expected Value is $20,000), and they paid us, say, $10,000 to use the software, and it cost us, say $2,000 to run the software for them, then the total value created was $18K, $10K of which went to them, $8K of which went to us.

This is very much a toy example, because it's hard to say what % chance they had of creating that deal without the software & the additional cost of adding that deal to the pipeline (in this case, cost was negligible, and probability of getting it otherwise was explicitly low).

Alternatively, you could look at value created from a time saving perspective: if finding the information WhiteWhale produces would've taken 10 man hours a week per user and you're supporting 3 users on a $200/mo plan, that's 120 hours a month saved. If their average hourly rate as a seller is even equivalent to $30, that's $3600 of hourly wages saved for the employer; $3400 captured by the team, $160 captured by us (assume 20% margins), and $40 over to our vendors.

So, in theory, if someone buys and uses our product, more 'work' is done substantially more efficiently and our customer's grow faster and the economy gets bigger and overall wealthier. Value is created!

[Value Creation != Value Capture]

Exactly how much value creation is a hard thing to pin down, especially because it's not simply related to how much money you make.

I'll give the canonical Peter Thiel example: Airlines create immense value, likely more than Facebook (would you rather get rid of Facebook or planes?*). You'd be misled if you used profits as a pure measure for 'value', though, because Facebook is insanely profitable, while airlines seem to be forced to either become a credit card company or go bankrupt.

Going back to the WhiteWhale example, we'd be capturing a smaller percentage of value created in the second example than in the first example. But, maybe that wouldn't matter to us if in the first case, we needed a sales team to justify & remind the client of that value created and in the second case was so obvious and low cost we didn't need to talk to anyone ever.

Going back to Facebook, we would observe a lot of their revenue coming from ads. The companies buying these ads would likely be getting positive ROI on them or they wouldn't otherwise keep buying them. So, maybe on the axis of advertisers, Meta still is creating value, even if capturing much more of it than an airline could dream of.

_{*If the answer is somehow planes, we've strayed further than god than I feared}

[Value Destruction]

Just like it's possible to build a business that creates value by letting someone trade money for time back, it's also possible to build a business that destroys value, too. If the pre requisite for being worthy of investment is value creation, then value destroying businesses should be avoided like the plague.

Note how I was talking about the advertising side of Meta / Facebook. If we go to the consumer side, you could easily argue that Facebook DESTROYS value.

We've been through this before: I'd never argue that social media is useless. Hell, it's how my busienss gets all of it's customers.

I would speculate, however, that on average, people's lives are net worse by using social media like Facebook. I'll spare you the full rant about algorithms designed to be addictive, etc, etc.

Just briefly think about the trade being made by using social media: the consumer gives their most precious resource, time, in exchange for a stream of lights and colors and sounds to give them dopamine while incidentally rewiring their brains to be less capable of valuable things like focus and presence.

And if you don't believe me on social media as something that is Destructive for the average user, just think about something else like Vapes that make people feel good in the short term but also make good progress towards killing them.

[Don't Invest in my Business]

One more negative example before we get to the positive ones, and this is a funny one, because it's an abbreviated version of why venture dollars shouldn't go to a business like mine.

In short, Jack & I are just building a useful software product. Yeah, I can tell you the long term vision of making markets other than financial ones more efficient, and that sounds investable, but I don't think you really need millions of dollars to work towards that.

Building a software business is hard to do, but the barrier to entry is kind of low. You can start building the thing today and selling it tomorrow and then rebuilding it the next day and reselling it the day after that. It will take you a while til you get there and a lot of trial and error and it's not for every one and you won't make any money at the start.

But you can keep going, and as long as you can avoid starving until you're ramen profitable*, you'll be fine.

You can build a software product that creates value without needing tens of millions of dollars to get there. Sure, there are reasons to invest in software companies and good uses of capital if you are a software company, but for the business Jack & I are building in particular, I still don't think it's the case.

_{*see: why I lived at home with my parents for a year and Jack kept his job}

[Old AI on Hidden Data]

A really cool use of digital technology that I've seen actually comes from a founder who lives at C House with me.

Basically, he built a product that goes into the pipes at factories producing liquid goods. This device lets them clean the pipes with a precision automation that minimizes chemical usage & downtime.

The alternative is having a guy kind of vibe check how much chemicals to use and generally have to wait longer than needed to make sure it's clean before turning it back on because he can't really tell.

Of course, once this data is collected by the hardware, you can do some pretty crazy optimizations & experimentation with AI to really figure out the cheapest and least invasive ways to clean. And, I don’t think you don't need an LLM for that, either.

This is really, really cool, because it's taking a real and important piece of hardware and using digital technology to unlock new levels of performance.

[Fighting Surveillance States]

I don't think I've written about surveillance states and fascism before, but I really do think that they are a great threat to an open society & scientific progress. Worth an essay on it's own.

China is already a surveillance state, and the US in some ways is & increasingly has the infrastructure in place to become one.

I think it's harder to roll back the clock on tech that's already out there than it is to produce and distribute counter measures. One such counter measure is a device like this that prevents all those annoying always on recorders from picking up your voice. (also founded by a former C House resident)

This falls into the same category as the above because while it's very much a hardware product there is also sophisticated software involved in making it work.

I hope I’ve gotten the point across that these two things have a strong interplay, hardware and software, physical and digital, atom and bit.

[Sci-Fi Maxxing]

Okay, I gave you the two real, grounded examples. Now we're about to go off the deep end into cool shit that I think would be awesome if it existed. Some notes:

• Not all of this requires digital technology, but I think in all cases it would likely help.

• These are just vague, broad ideas. You still need a way to make money & really good execution (sorry JL).

• I’m very technically naive about all of them

• I say these are potentially interesting spots for VC investment because this is where putting capital in the hands of a shrewd operator would make a difference in the outcome of the venture

[Nuclear]

I wrote my college essay to Ross about putting nuclear reactors in places like abandoned coal mines in Appalachia. One of the graders thought it was cool and ambitious, the other one thought it was basically brain dead.

Nuclear is one of those technologies that from an admittedly naive perspective seems that if we were to aggressively invest in, would lead us a lot closer to hyper abundance.

It’s also one of those technologies that seems to have been poisoned by regulation into irrelevance.

[Space]

I, like perhaps 20% or the US population, wanted to be an astronaut when I was younger. Unlike that 20%, I never actually gave it up. I have a to do list from when I was ~19 with ‘go to space on it’. Given that I’m already about 1/3rd through the to do list, I’m never giving that item up.

There's an argument that going to space is wasteful because there are problems to solve on Earth. Here are my 2 biggest counters to that argument:

1. The technology it would take us to colonize the solar system and then the galaxy would be incredibly valuable to life on Earth: self sufficient food systems, robust 3D printing, super efficient energy systems

2. The problems on Earth are irrelevant in the event of some global catastrophe that we can't contain, whether it be man made or astronomical like an asteroid wrecking our planet. We don't know if life exists elsewhere. If we think life itself is valuable, then we should be invested in protecting it from both ourselves and chance; an interplanetary (or better, interstellar) society is a massive step in that direction.

And, it'd also be really cool to go to space.

[Vertical Farms & Regenerative Agriculture]

Why aren't our city scapes covered in greens hanging from windows everywhere? I feel like I saw some movie with a futuristic city scape integrated with microgreens everywhere when I was a kid but it never happened. Why not? I don't know. Maybe there's a good reason, but probably not.

One of the engineers I respect most took an indoor vertical farming company from 0 to unicorn status. More stuff like that would be awesome.

And, while we're at it, I think that regenerative agriculture is far better for the environment and human diets than monoculture; worth exploring how to to scale it up as much as possible.

The empire state building but covered in these

[Bio & Longevity]

I don't know shit about bio, but man it would be cool if we cured cancer.

And, while we're at it, why not increase average lifespans from whatever they are now to like 150 years or something. I met a guy working at a startup to grow organs for replacement. According to him, that gets us quite far, especially if you can also find a way to increase the health-span of the brain.

If you love hearing the weekly ramblings of your friendly neighborhood bootstrapped SaaS founder, smash that subscribe button.

[Read SciFi]

After writing this, I can see the temptation of wanting to be a VC. You just get to talk about & invest in cool stuff.

I can also see how it'd be easy to get fucking scammed if you did't know what you're talking about. VCs get scammed so much that sometimes I have to wonder if they like it.

That being said, I can hope the impending contraction is like a forest fire that removes the blight and lets healthy, strong, & wise firms grow.

The future tends to look better when we invest in and cultivate technology that genuinely creates value. I can understand if you’re disillusioned by the recent uses of capital to reinvest in things that aren’t really making anything better.

That’s not a call to go inward and be afraid of what’s next, that’s a call to remember the things that actually give us hope and make the world more exciting and awe inspiring.

To be clear, I'll likely work on something from the above list once I achieve my financial ambitions with WhiteWhale.

Dream big.

Live Deeply,

CLVI

[The Antithesis of the Bubble; PLG - Week One; The Code Base of Theseus; It's Still Iterative; Not To Burst Your Bubble...; Learned Helplessness; Fear & Loathing]

Thesis: Building good things is hard; when the bubble bursts, don’t mistake the fallout for evidence that technology is bad.

[The Antithesis of the Bubble]

This post is a bit disjoint because I first follow up with some exciting but very early results on WhiteWhale's swap to Product Led Growth (PLG) from last week, and then I dive back into the AI bubble.

Still, the two are connected in a few ways.

In one sense, we are using the leverage baked into a PLG motion to bet against the AI Bubble.

More than that, though, I think the process we've been using to iterate on WhiteWhale is a really good contrast for the way that the loudest people in the AI bubble are implying or outright saying that things get built.

This is my biggest gripe with the AI bubble: how thoroughly AI Labs & the media & generally AI boosters seem to have convinced everyone that LLMs are obsolescing people & skills.

No matter what they've told you about the death of software engineering or the end of this skill or that, building cool things is still really, really hard & takes focused time and effort.

Dario & Altman are finally ostensibly rolling back on their fear mongering everyone-is-losing-their-jobs-claims, but this is only after or in conjunction with companies like Uber admitting they are struggling to find ROI on the ai spend.

I'll explain both why it's too little, too late, and why their should be no quarter given for the lies and fantasies spewed over the last few years about what AI maybe will one day possibly do.

I'll also clarifying how this stance of mine is coherent and actually amplified by another, seemingly contradictory claim: I adore technological progress and think AI is a core technology we should continue exploring.

[PLG - Week One]

Product Led Growth (PLG) is going unreasonably well for WhiteWhale.

In the first 8 days of having a buy now button on our website, we had two customers buy without speaking to us, plus a third purchaser who I demo'd a year ago and dm'd when I saw him in the flow.

I say this is unreasonable because the volume of people who have clicked on our flow is hardly significant.

I hesitate to share this because I generally don’t like sharing numbers, but what the hell: in 8 days of launching, we added $900 MRR from the PLG motion with 40 people going through on onboarding flow & giving no free trial.

For a VC backed company, this is small potatoes; but for two guys with the goal of hitting $1M ARR without hiring anyone or raising any money, this is a lot better of a start than I thought we’d get with PLG.

Regardless, this first week could literally just be a fluke, and there are other levers we need to pull to make it work in the long run. I’m super grateful for the start, though.

[The Code Base of Theseus]

I speak about iterative engineering a lot (eg, here & here). Meaning, you start with one solution and then, overtime, you work your way towards a 'more optimal' solution

Take the onboarding flow we just wrote. After I had the simplest first draft I could ship, I've already gone through it nearly 70 times to fix bugs and iterate on the experience. Perhaps 50 of those were before we put the button on the website; and Jack went through it, too.

Still, we both missed things that became obvious once we started watching user sessions*.

Copying the RB2B flow, we have our users add our bot to their Slack in the onboarding flow itself. As part of this, they need to pick which slack channel they want to add it to.

When I was testing this, I was doing it with a dummy slack I made for these purposes. That means, there were only 3 channels in this slack, so I could find the right one from a list very quickly.

But, when I was watching recordings of real users, I quickly realized that most people have a TON of channels in their real slacks (we do, too!).

So, in practice, I realized we were missing two bits of functionality on this page that neither myself nor an LLM picked up:

1. Freezing the refresh button to pull in new channels they might make at the top of the element; otherwise they could scroll down, make a new channel back in slack, and because of the hidden refresh button, not realize they needed to click refresh to find the channel

2. Adding a search bar for the slack channels, frozen at the top next to the refresh button. Otherwise, you might have to scroll through 50 channels to find the right one!

Maybe if you're actually a good front end engineer (truly, I'm a novice), you'd just 'know' you should do these things when designing. But, for the rest of us, just shipping and watching is a super power. Users will pick up things you miss.

The same about iterative design goes on the backend: we had some performance issues this week because the backend was putting too much load onto the database.

So, I played the game of finding the most resource hungry query across all sessions and making changes until it was not the most resource hungry query.

I did this across 3 queries and severely reduced the load on the database by an order of magnitude from peak to new average, and by a factor of 3 or 4 from old average to new average.

Again, maybe a better engineer would have done this 'right' the first time around, but one of these changes involved creating another table. In other words, it was a heavier fix that I was able to make more clearly with a lot of information I gathered from the time I originally wrote it. I do think in this case, even a more competent engineer would not have seen this right away; even if they did, they wouldn’t know if it was the ‘right’ call.

Codebases are like the Ship of Theseus: you replace one part at a time until it's no longer what it once was. Unlike the Ship of Theseus, each replacement is not just maintenance, but also often an improvement.

_{*Being able to watch the real playback of a user session is insanely valuable, by the way. This is not a paid ad, I'm literally still on their free tier, but I am in love with Posthog - it took literally 30 seconds to set up. Perhaps I'm exaggerating, but it can't have been more than 5 minutes.}

[It's Still Iterative]

I share all of this because, for one, the engineering process is deeply interesting to me.

The second reason, though, is to remind you that the engineering process really is a process.

You discover more information over time as people use your contraption, and as it interacts with the world.

An LLM cannot 'see the future.' Perhaps it can come up with a better solution on average than a novice in some fields, and it can often do this faster, too.

This doesn't inherently mean it can get you to the end state faster, though. There is still a process that takes time and iteration.

And if you let the LLM produce a bunch of code you can't understand, it can actually slow you down and create structures and patterns that simply don't make sense or are about 490,000 lines too long but now have to be maintained by humans, who can't or simply don't want to do this now overwhelming jobs without LLMs!

And to be clear, for my front end, I heavily use LLMs for the CSS / HTML / JS. It's not like the LLM magically caught the slack channel issue above, nor would I expect it to, just as I wouldn't even expect myself to.

The lie that software engineering is dead or LLMs can do it all is simply that: a lie.

Sure, it can spit out some code that kind of works, but so can an underpaid intern. It's not done yet, just like I don't think a competent developer who care about their output would ever think their first pass was done, ever.

'Done' comes in a gradient and you get a lot closer if you go through it 50 times before you launch it, with an obsessive attention to detail. Even then, you will miss things.

[Not To Burst Your Bubble...]

_{Disclosure: In May, I put my money where my mouth is and took out a position against Oracle. I made this trade because I think putting more skin in the game on something I talk about so much aligns my incentives more than not trading does; I lose money if I'm wrong, not just credibility. Unfortunately, this also means I can lose money if I'm right but place a sloppy trade; I'll likely double down on it next week to fix that.}

Ed Zitron wrote this week that he thinks we're near the end of the AI Bubble. I do, too.

Some of the executive driving AI spend are starting to realize they're more or less burning piles of money on dubious to 0 roi.

Anthropic & OpenAI have raised so much money, it's not obvious where any more capital can come from barring the exit liquidity provided by public markets.

Elon is actually a role model of mine in number of ways, and I really like the fact that the richest guy right now has been the driving force behind building real technology that has changed the world. But man, I'm not a fan of SpaceX's impending rushed inclusion in the NASDAQ-100 at all.* But hey, I guess I'd rather have people throwing money at a company that I believe will actually have a long term positive impact on the world instead Anthropic & OpenAI.

After all, Antrhopic & OpenAI, are deeply unprofitable & fundamentally flawed businesses kept afloat by massive quarterly capital injections with no path to profitability. And no, some leaked deck from a vc that vaguely says ‘they'll have as much revenue as Microsoft' is NOT a path to profitability. It's just more marketing.

These companies make me really, really mad, because they've spent the last 3 or 4 years fear mongering about how people would lose their jobs and driving this obscene wave of doubt around the value of learning real skills.

I don't care that Sam Altman and Dario are rolling back on this nonsense now, that doesn't make up for the last 3 or 4 years.

Self explanatory, I hope?

I can't tell you how many times I've spoken to people who were genuinely concerned they weren't needed anymore and their skills were being obscolesced and it wasn't worth it to learn things vaguely because AI will replace them and do it all and that all that's left for humans is relegation to some ambiguous and ill defined skill like "taste".

_{*In short, some rules were changed that allowed SpaceX to be included in a major index very rapidly upon listing. This means that the stock price gets additional support by purchases made by large etfs and index funds. If you have a 'vanilla' financial advisor or retirement fund, there is an incredibly high chance that you will soon be an owner of some shares of SpaceX without realizing it.}

[Learned Helplessness]

Part of the reason I focused so much on the iterative engineering process above was to contrast it with some of the things I've seen in the past few moons among the kind of person who was sold the narrative that if they downloaded Claude Code salvation would be theirs.

I had one session where I watched a person attempt to use Claude Code to install and run a git repo for about 30 minutes. For context, there were 15 minutes before this where the person was not actually signed into Claude Code and therefore couldn't figure out why nothing was working.

The most painful thing about this whole endeavor was not a non technical person trying to do something technical; that is actually a beautiful thing, in my opinion - everyone starts as a beginner in everything.

What was painful was a sort of learned helplessness I observed in this person. There was no attempt to conceptualize what he was doing start to finish, no planning. There didn't seem to be a conception of where he was in the process, either. When he had to restart Claude Code and it came back and said "I don't see the thing you asked for," he didn't even try to give it a history of what it had just done; I found myself dictating one sentence prompts as simple as "You just said you installed this thing last session, where is it?"

LLMs have offered to teach helplessness en masse and so many people got in line to learn it.

"Oh, you don't need to be spending your time on that, here, let me do it for you!"

My brother in Christ, sometimes you do need to spend your time on that. It's the only way you'll learn and get better.

Fumbling around in a GUI disguised as a terminal will not solve your problems or make your life better.*

I've said it before and I'll say it again: LLMs really can be useful! I use them to help with coding often, particularly my front end. But, as you may be able to guess with the process described above, even using them to help, it took me well over 100 hours to rewrite our front end from scratch and get it to a place that I felt was shippable. Since we started rolling it out in mid May, I wouldn't be surprised if I'm approaching another 100 hours spent on it!

Building cool things is hard and takes effort!

_{*I still can't get over the fact that a 'terminal' application was written in fucking react & has taken over a year to maybe kind of solve a screen flickering issue by introducing a ‘no flicker mode’ that sometimes works after failing with an 80% fix that made other things worse. To put this in context for non technical people, this might be like someone whose never mowed the lawn or trimmed a hedge trying to trim a hedge with a lawn mower, cutting off their foot, and then tweeting, 'Hey, I'm still 50% able to stand and the hedge looks a little better, so this is a win!’ only to later introduce ‘no amputation mode’ as if it were a fucking feature.}

If you enjoyed this post, feel free to subscribe, or don’t. I’ll be here each weekend regardless.

[Fear & Loathing]

I've written about the AI bubble a lot, and these last few weeks I felt like I didn't need to write about it as much anymore because we really are likely near the end.

While I understand that a lot of people hate AI now, I hope it's for the right reason.

I really don't think you should hate OpenAI and Anthropic because of some vague fear that they'll steal your jobs and drive unemployment. Unfortunately, this was part of the con. There are plenty of other reasons the layoffs happened, including overhiring and an economic downturn.

I am actually a huge proponent of continuing to build and grow towards Artificial Intelligence in it's myriad forms, probably even cybernetics.

What I am upset about is that the market was sold a lie about what specifically LLMs could possibly one day maybe do and they bought it like it was the truth today. Something like a trillion dollars was dumped into what I believe to frankly be a dead end AI research path when there are a million other things we can do to further understand & augment intelligence & build useful things.

Expensive GPUs are tirelessly laboring away in data centers to write emails no one will read and summarize emails no one paid attention too and probably didn't need to anyways.

People were told AI was scary and dangerous. It gave governments another fucking thing to waste billions of dollars on talking about maybe regulating. Think about how many meetings happened in Washington DC discussing what needed to be done about a fucking token prediction engine, how many c suite or board meetings started with someone so far removed from the actual fundamental business that they thought the best use of everyone's time was to make the whole fucking meeting about the "AI Strategy".

Obviously, OpenAI & Antrhopic are not the only ones to blame for this - if our society didn't have such a strong disposition for bullshit, no one would've bought these narratives to begin with.

Again, technological progress is fucking amazing.

It put us on the moon, let us harness the power of stars, lets me send this blog at the click of a button to hundreds of readers, allows my startup to process terabytes of data to find valuable information at a cadence and accuracy that you'd need hundreds of human researchers for, lets us grow food in our caves of steel, transplant hearts, delays or stops the onset of crippling diseases, and create aluminum beasts that take us from point a to point b with no human driver.

But Jesus Christ, why'd we have to waste a trillion dollars on fear mongering campaigns and fucking summarizing emails & writing a bunch of slop code that no one will use?

I fear that the fallout of this bubble will be bad for science and progress. I worry that the narrative will not be, 'decision makers acted like children who got into a strangers van because he offered them candy, shame on them for being so willfully blind to reality and such terrible capital allocators.' I worry that the narrative will land closer to fear and loathing towards wealth and technology.

These really are two categorically different things; the former points us towards some potential incremental improvements and changes while the latter leads us away from the stuff that drive progress.

I cannot stress enough: technology, AI, and investment in the future are not bad.

What's bad is selling & buying a narrative that this one piece of technology with no provable ROI is this dark and dangerous power that can concurrently jepoardizes and can save the world. It’s a con and a power grab.

Please, remember, always: there are no silver bullets.

Live Deeply,

CLV

[A Structural Change, Boldly; The Fuck You Money of SaaS; The Money Burning Alternative; Defining Risk; The Upside; XP Farming; Procrastinating Away Anxiety; Be Bold]

Thesis: Stop tweaking and take the risk

[A Structural Change, Boldly]

Dear Reader, the deed has been done.

I'd be lying if I didn't say that it gave me some anxiety, but that's to be expected.

It might not seem like a big deal to you, and it really, in the grand scheme of things, isn't a big deal at all.

But it's a big deal precisely because it isn't a big deal when it should be a big deal.

Jack & I just flipped a switch on the business we spent the last 2 years building into something that can support both of us.

In the past, if you wanted to buy WhiteWhale, you'd have to talk to one of us on a call. Now, if you want to buy WhiteWhale, you can do it from our website without speaking with us.

This is a switch from what was "Sales Led Growth" to now being "Product Led Growth" (PLG).

It's a big deal because it changes the levers we need to pull to hit $1M ARR with just 2 guys. And, I think it's one of a few changes we could've made that make the goal actually possible.

The beautiful thing is, if we fuck up and are totally wrong - well, we can just change what we're doing again. Today, I want to talk about:

• Our bold goal & why we're pursuing it

• How the decision to switch to PLG is actually close to riskless

• A few notes about ignoring undue anxiety

[The Fuck You Money of SaaS]

First off, the goal that is driving Jack & I’s decision to switch to PLG is a very aggressive one: We want to build WhiteWhale to $1M in Annual Recurring Revenue (ARR) without either taking on any outside funding or hiring other people.

This is a pretty unconventional goal, because raising money and hiring people are two big levers you can pull to help your business grow, and we’re actively saying no to both of them.

I know a few founders who have done things more impressive than this (think 2 founders + 20 VAs doing $20M in revenue in 2 years), but accepting both of these constraints really isn't so common.

To put this into reference, here is a table of some real examples of numbers from founders I know or know of who either I or someone else would call successful:

If you couldn’t guess, the very last row is the most attractive one to Jack and I. The guy who achieved it, Adam Robinson, calls it the "Fuck You Money of SaaS". So, we do know that it’s very, very possible.

[The Money Burning Alternative]

To be clear, I’m not objectively against raising money or hiring people overall. I think the founders behind the second to last row will genuinely change the world and IPO in 5 years.

That being said, given our space in particular, and the outcome we want, I think it makes sense to not raise and to not hire. We’ve seen the alternative in our space.

One company, as an example, raised $50M total, sold to an incumbent, and word on the street is that the 'acquisition' was for a fat $0. (this is one of the ostensible benefits of raising enough money; even if you fail, you can find someone to 'buy' you for $0 to publicly save face)

When you start raising money in our space, you tend to instantly have aggressive growth targets that you need to hit. If you don't have Product Market Fit (PMF) yet, basically what you start to do is hire a sales team to force annual contracts down your customer's throats and justify them by adding half baked features that aren't really part of your value prop. Ultimately, you lose focus on finding PMF and chase revenue that is only sticky because of your sales team + terms.*

And, on top of that, I just genuinely don’t know if a lot of our competitors can ever justify the valuations they’re achieving. Outside of the CRMs, the ‘best outcomes’ in sales tooling are rather grim by venture standards: Clay, the rising star, is at $100M ARR and has a $5B valuation. To put that into perspective, ZoomInfo, one the biggest established non crm players in the space, has something like $1.25B in revenue and is valued at ~$900M.**

And, more than all that, Jack & I know the outcome we both want. We're both workaholics, but we're fighting for freedom, not to work more. We both have a laundry list of other things we want to do, and manage a bunch of employees or get in arguments with a board of directors is not on the list.

Okay, but why not just hire an employee? Because we want the leverage to be inherently baked into the business. And, as a I wrote about last week, baking it into the product is a more pure play way to do that.

This is why we switched to PLG; it solves for a structural bottle neck that would otherwise need to be solved by hiring human capital.

_{*If the alleged ‘SaaSpocolypse’ is real, I really think it's because investors were not appreciating the problems with this model, not because of LLMs and pixie dust.}

_{**If you don’t understand why these numbers are bad, depending on the financing terms, if you raise $50M like the competitor I mentioned did, the bet your investors probably made is that you’ll be worth at least over $1B, if not more! If the biggest provider in the space, ZoomInfo, isn’t even worth that much according to public markets, the likelihood you can find someone who believes your private business is worth way more than that to fund you further or by your company goes down a lot.}

[Defining Risk]

When you have a vague fear that something could go wrong, it's easy to make a big deal about some vague and hairy conception of risk.

But getting down to the nuts and bolts and really understanding what ACTUALLY can go wrong and how bad it could be when it does is key.

Since we've bootstrapped WhiteWhale, we are blessed to have a profitable business with basically no liabilities. In this sense, if we stopped getting new revenue for the next month while we played with PLG, we're totally fine.

This is one of the great blessings of the way we're building: We have no burn rate, no ticking time bomb that says you need X new revenue by this date so you can raise more money or you'll go out of business. So, if we're dead in the water for a month, and this PLG stuff doesn't work... well, we can just roll it back!

On the other hand, if we raised capital, we would have time tables to keep up with, and risking one month of new revenue could be the difference between life and death.

There is nothing stopping us from swapping the buy now button back for the "book a demo" or talk to sales button.

[The Upside]

And, on top of that, we were actually able to make the swap a ‘profitable’ trade even if we do end up giving up a month of revenue. After all, one of the pre requisites to make PLG work is a product that is self explanatory and easy to use. To get to the point where we even felt comfortable putting the buy now button on the website, we had to be super, super aggressive with the product design and features.

Everywhere we've seen where more than one user gets caught and confused, we've been deleting or redesigning or rebuilding or redoing.

This, overall, just makes the product better and reduces the number of users who ask us how to do x or y or z. And, it removes the number of times we actually have to do x or y or z because the platform feels like it should do x but doesn't. So now it either does x or it doesn't.

That's not at all to say that the product is done - it's definitely not! We're still doing this process very aggressively.

But already, in a lot of ways it's not recognizable compared to what we had 3 months ago.

As a concrete example, check out this before and after picture of our main dashboard:

Old

New

If you’re not a user, it might seem like some marginal differences, but we’ve removed unneeded pages and hidden menus, and importantly, our usage has gone up quite a bit, and support questions are going down.

Back to the point - if PLG fails miserably, it still caused us to make a product that is far better and easier to use than it ever was before.

[XP Farming]

If you're taking risk and paying attention, you should be learning.

Meaning, maybe you want to make a big move and place a big bet, but so fucking what?

We know about a million times more than we did 2 years ago when we were starting WhiteWhale (then BirdDog). It'd be outlandish to think we’d be less equipped to make this kind of decision than we were back then.

And, guess what? We can still be wrong! But, as discussed above, we've structured the bet to chop off the downside, so oh well.

If you're making some big decision, I'd encourage you to think about the same - what other big decisions have you already made? What did you learn? What can you do to control the risk of this decision?

[Procrastinating Away Anxiety]

One other source of anxiety in the face of bold decisions is general worry about all of these little commitments and other shit you have to do. But what about this and that and this?

Maybe you have some other things you're involved in that are making you say oh I shouldn't take that risk because of this this or that. I, as of late, have a couple of things like that.

But guess what? You'll always have items on your to do list that distract you from the big thing or goal you're trying to accomplish. Don't let them stop you. Again, actually map out the downside and handle them accordingly.

I trick I started to use is procrastinating as long as possible.*

What I mean by this is if I have some task I don't want to and is really low leverage but I can't get rid of it super easily, I'll figure out how long it will take me to do and the longest amount of time I can wait to do it and then schedule to do it in that much time close to the deadline.

Meaning, if it doesn't need to be done for 4 days and will only take 30 minutes, I won't let it disrupt my flow or focus now. Rather, I'll put it on my cal for 30 minutes 3 days from now and stop worrying about it.

It's not a problem til it's a problem, and you'll always have a to do list. There's no point in tweaking about it until you do it, just stop thinking about it instead and tackle it when you need to.

_{*As a word of caution, this advice is given in the context of being one of those type a over achieving people who tend to do too much, not too little. If you already procrastinate and don't do anything, maybe this advice doesn't apply to you, or maybe you've been wise beyond your years for your whole life.}

If you want to follow along on the journey to $1M ARR and me finding myself along the way and maybe helping protect against the fall of the west, please subscribe

[Be Bold]

Go out and be bold. Whatever you're worrying about probably isn't that deep.

If you have a crazy stupid ambitious goal, go for it.

I’m sure you can take the risk in a way that doesn't mortgage your house or whatever.

Be bold and handle that shit.

Live Deeply,

CLIV

[Code that's Lasted Longer than any of my Girlfriends; Data Monkeys; Eliminate, Automate, Delegate; Tech Enabled Service; Grinding to Valhalla; Leverage Via Omission]

Thesis: To get the leverage you seek, work on your business, not in your business.

[Code that's Lasted Longer than any of my Girlfriends]

I was sitting in my room last night with a friend. One of those conversations where I was 95% focused on her but was watching my terminal out of the corner of my eye to swat any bugs that popped up from a decently involved change I had just made to WhiteWhale*.

Enter: WhiteWhale

When there was a lull in the conversation, I had quite a holy shit moment: in one form or another, the code that I was watching had been running for almost two years now.

I started writing our code immediately when we launched in July 2024, and it's about to be June 2026. The first few months involved aggressive writing and rewriting. But there's been some continuity the entire time: some of the lines (and a lot of the philosophy) from 2024 remain untouched, even if so so much else has shifted about.

Funnily, I was actually running WhiteWhale out of Tmux windows until January 2025, and backend wise, we rode one server in my basement right passed $100K ARR.

Fondly reminiscing on code, I'm sure I'm reminding you that I'm a total fucking nerd, but that's okay.

I may have said it before, but it warrants restatement: even if I was as good of a developer as I am now back in July 2024, it would have been impossible for me to write WhiteWhale in it's current form then... competence matters, but so does all the time we've spent trying different things out with customers and deciphering what they actually need, rather than what they say they want.

The same goes for capitalization, too: if we had started with more than the $3,000 that we did, I don't know how much faster we could actually move. I'm sure there's a number greater than $3,000 that could’ve sped us up, but the money on it's own would not have been enough. It'd be just as likely (if not more) that we would have burnt that money in a funeral pyre of inefficiency and moved even slower.

This week, a mentor said a phrase that stuck with me:

That they do. And a bootstrapped startup and a well funded one likely will take the same amount of time to find PMF… but both are more likely than not to miscarry and never find it at all!

This post is all about leverage, as so many things are. Particularly, how Jack & I have and are continuing to use code to build leverage into what started as a service business.

This is what we have to do if we want to get to particular revenue targets without hiring another soul. It also means being guarded against spending too much time running the business while still not optimizing it too soon. And, it's also why we're taking the final grand steps to Product Led Growth (PLG).

_{*Yes, it happened - we rebranded from BirdDog to WhiteWhale!}

[Data Monkeys]

Prior to starting WhiteWhale proper, Jack & I got about 6 months of experience speaking with sales people and manually delivering research services to them.

This was super valuable in what it taught us about our space, but it would have been an incredibly low leverage business if we tried to scale it.

The engagements were all a bit different, but they looked something like this:

• Sales team gives us a list of organizations they can sell to, say US based police dispatches

• We very manually did aggressive google searches and read meeting minutes trying to find trigger events, like quotes from leadership, grants, specific technology investments, open job roles, etc

• We would summarize the information in a little narrative supported by the data, saying target this org now because of these triggers (signals)

• This was delivered either as literal PDFs or in a low tech front end that was basically an interactive pdf

We would each spend over 40 hours a week manually delivering this to our clients, running around between a bunch of different sites and inputting the data into a spreadsheet with some particularly finicky automations.

It was very, very slow going. The business was capped unless we hired other people.

If we wanted to get bigger, the bulk of our costs would go up linearly with the business - if we wanted to add 2 more customers, we might have to hire 1 whole person! And even if this person was a VA from the Philippines, we would still require training and infrastructure and management to make it work. Meaning, hiring people would also increase the complexity of our business, too.

And, even if we hired a lot of people, it would be impossible to deliver the value we do now without a very large army of employees. Automatically, we now check tens of thousands of accounts for trigger events and other signals every other day! Good luck getting humans to efficiently do that!

[Eliminate, Automate, Delegate]

Before we continue along the WhiteWhale journey, I want to address a common question we get: why don’t we hire or outsource this or that?

There's two adjacent pieces of wisdom that give us pause about hiring:

This latter quote gives us an order of operations to follow when improving a systems:

1. First, Eliminate

2. Second, Automate anything you can't Eliminate

3. Third, Streamline (Optimize) & Delegate anything you can't automate*

Delegation and optimization are more expensive in the long run than either automation or elimination. And, both require upfront resources and time to get right.

So, yeah, I guess we could've hired a commission only sales person as early as November when ACV & win rate were both high enough to do so.

But, doing this is the lowest leverage & most expensive option in the stack - we'd need to first systematize a process and give it to them.

If you do that too early, you're optimizing something that might change rapidly. We didn't know we would go to PLG in November, but we did know something was still 'off' that we needed to solve. And we weren't ready to invest the time and effort into delegating something that might rapidly change later.

And now, we're convinced we can eliminate sales as a requirement to get new customers altogether.

All of this said, I must admit that Jack & I are on the slower end of streamlining & optimizing systems, and probably also slower on the automation bit, as well. This is a style of building that bootstrapping affords for us, and perhaps even neccesitates.

I’m sure we could be faster to automate & streamline, but I think it works out because we're a lot more trigger happy with eliminating things. Which, again, is the highest leverage option you can pick.

_{*Roping Streamlining & Delegation together is my personal view of the matter, as the goal is to bake leverage into the business. Streamlining doesn't give you as much leverage since it doesn't remove yourself from the task, and delegation is made easier by streamlining. Although, I do believe the 'last job' of a CEO is as the portfolio manager / investor.}

[Tech Enabled Service]

After the many ups and downs of building and selling software, at some point late last year, we realized that WhiteWhale was basically a tech enabled services business.

What I mean by that, is we had a piece of software that worked and provided value. This was a massive improvement, as we became removed from manually combing through the internet, looking for all of those articles and trigger events and signals. Now, our software was & is doing this across tens of thousands of our client's target accounts each night and automatically updating itself and presenting the information to them.

I don't know if you can appreciate how much this feels like magic until you use a piece of software that, at the press of a few buttons, does something that took you hours or tens of hours a week to do yourself.

However, a few very big things still required input from Jack & I:

1. Selling the software

2. Configuring & managing the software (defining & helping update the triggers and signal events)

3. Reselling / Upselling / Customer Success functions

While we made it out of the grindy, data monkey trenches, we found ourselves in a spot where without our continued input, the business would fail.

To be clear, this is not a 'bad' or untenable situation, but we had basically given ourselves jobs. As I wrote about a couple weeks ago, since we are our own bosses, this still puts us far closer to fuck you money than if we were making the same amount or even a little more at someone else's company. And, it's actually a sustainable situation, unlike our data monkey hustle of years passed.

But, the business was bounded both by our ability to get on sales calls with people AND our ability to properly onboard and support them after they bought. That’s not what we want!

We put a ceiling on ourselves. And, we impeded customer acquisition, too: because our time was part of what we were selling, we had to raise the minimum price far higher than the software neccesitated on it's own. This obviously lowered the number of people who would buy from us.

So, going back to our options of Eliminating, Automating, and Delegating, we're in the process of eliminating our sales function, automating the configuration and all other inputs we have to give to help someone get ROI on the software, and then doing a mix of eliminating / automating / systematizing & maybe delegating for the last bucket of items.

Altogether, this is & will give us more leverage than we've ever had.

[Grinding to Valhalla]

In practice, this looks like aggressively making a million changes that will keep moving us out of the loop while making customers happier and happier

• We got one of our customers to jump their commitment up from one quarter to 3 quarters with an automated email our system sent out about how WhiteWhale impacted their pipeline (showing ROI)

• We entirely rewrote our front end from scratch; it was on a no code tool, and now the new edition is not, and there is no Javascript framework; overall, it means we can iterate on it a lot faster than we could before.

• We wrote down everything we've learned about making good signals for customers, and built it into a system that does it on it's own.

• Our customers can upgrade their account within the platform without talking to us

• We added a wrapper over our signal input system that used to require users memorize a specific syntax, but now just accepts free text.

• The new front end also has a more intuitive to use account finder that mixes right in with the user's regular accounts.

• We’re adding a human readable log to the account searches that will eventually have recommended adjustments so users can understand & get better results

• Jack rewrote our website to have a much stronger funnel full of video customer testimonials from support calls.

• We’re finishing an onboarding flow that can take a user from nothing to having a configured account with signals and sample data before prompting them to buy without talking to us.

• We will re-document the new platform with videos & text

• We will enroll users who complete the onboarding flow in a sequence that remarkets those who didn’t buy and reminds those who do of the value of WhiteWhale and how to get it.

• We're gathering better telemetry that allows us to see where users get stuck or make 'mistakes' in the platform so we can code it out.

• At some point, we'll add in an AI CS tool to start resolving the low level support issues that you can never fully engineer away.

It's easy to get so caught up in running a business that you forget you’re building a business. This is perhaps one of my greatest shortcomings as an entrepreneur. But, we’re addressing it now.

If you enjoyed this post, subscribe - I’m here every sunday writing about whatever is on my mind, from the AI bubble to bullshit work.

[Leverage Via Omission]

There is a lot of hard work between where WhiteWhale is now and getting it to where Jack & I want it to be. But, there’s also a ton of hard work that's already happened between the time we were manually grinding on research and where we are now.

It's pretty cool that some of the code has been with us the whole way, and works for us without sleeping, without resting. That's a beautiful piece of leverage.

The most beautiful piece of leverage, though, is everything that's not there, all that we've already eliminated, and all that is to be eliminated.

Live Deeply,

CLIII

[An Allergy to Bullshit Work; Petition to Move an Exam; Skin in the Game; Identifying Real Work; You'll be Wrong; Don’t Create More Work; AI Slop Bombs; Do the Hard Thing]

Thesis: Avoid bullshit work like the plague and do the few things that will actually move the needle.

[An Allergy to Bullshit Work]

It is way easier to do bullshit work than real work.

But, if you are able to avoid bullshit work and just do real work, you will achieve more and be more fulfilled, too.

I'm writing this post this week because I recently got a few messages that on the surface 'seemed' productive to the sender but really just magically generated more bullshit work than was needed to achieve the desired outcome.

I'm taking the time to write about this because while creating bullshit work for other people is no way to live, sometimes the differences between actions that create this fake work and ones that don't can be subtle. You might genuinely not realize it, especially if you operate in an environment where other people do this all the time.

It might seem harsh, but I would argue that if you’re aware of what you’re doing, creating bullshit work is a selfish and antisocial act: you're wasting someone's time that could be spent on something meaningful.

This is particularly impactful in environments with very aggressive & honest feedback loops, like startups, where avoiding bullshit work and doing the high leverage items can be the difference between succeeding and failing.

So, coming up in this blog post:

• We revisit my college experience in a den of business LARPers

• We talk about what actually constitutes real & fake work and how to identify the difference

• I complain about some specific examples of people creating more work for me in ways that should be criminal

[Petition to Move an Exam]

Man, I loved the University of Michigan as a place with people at it.

But Ross, the business school, was such an attractor of people who love bullshit.

Ross, at the time, was the 3rd best undergraduate business school in the country, second only to Wharton and either Haas or Stern (can't recall). Yet it was FULL of people who thrived doing bullshit work. They spent far more time signaling that they were doing work and talking about how hard class was than they spent studying or actually doing anything.

We'll call this type of person Ross Natives, although there was another, more vulgar term on campus we could perhaps apply to them...

They got so invested in and caught up in the narrative that they were God's chosen few who would walk out of school with a job paying between $80K-$120K and storm America's C Suite over the next 15 years. To them, this narrative was gospel, they lived in it.

Some tell tale signs of Ross Natives:

• They'd complain about how hard classes were all the time

• They'd always 'study' in groups in public places (the beloved Winter Garden) so you knew they had friends or were part of the 'in group', but obviously no studying was done

• A large part of their identity was the clubs they were in and then the internship and then the job they got

• They 'studied' by memorizing practice exams rather than by internalizing any of the materials or lessons

The important pieces of this are that they tended to avoid the real work of learning and replace it with time signalling and making the universe think they were working.

To paint the picture with the best example I have: In one particular instance, our entire class of ~550 had an accounting exam coming up. For whatever reason, some of the students, certainly all Natives, decided that the academic calender the accounting department had planned out wasn't actually fair or equitable or some other such nonsense. (the real 'reason' they gave is escaping me)

Regardless, this group wrote a petition to move the exam back one week and got it signed by enough students that the accounting department listened! And they did so in what I remembering being a weirdly apologetic way, saying, "You're right, we're sorry, we didn't think about that."

Writing, signing, and presenting this petition falls squarely into the category of bullshit work. If whoever contrived this scheme or was party to it would have spent the same amount of time they spent on the endeavor actually studying, they would've been fine for the exam.

Instead, they created more work for everybody: - themselves for writing it, admin who had to respond to their emails, the accounting department for having to shift the schedule, the students who actually studied and would now have to stay fresh on the material a week later.

This is why I posit bullshit work is selfish: it self propagates like a swarm of rabbits in spring. These students couldn't be bothered to study for whatever reason, so they did some bullshit work that created more bullshit work and wasted everyone’s time.

The most mind numbing part of the whole thing is that Ross is graded on a quite odd curve: while it stopped more than 40% of students from getting A's in a given class, to get lower than a B, you basically had to have beef with the professor.

At best, the people who wrote and executed the petition shifted some As to Bs and some Bs to As.

[Skin in the Game]

In a troubling turn, reality will not disprove the narrative these Ross Natives have in their heads. Many of them genuinely will end up as partners at consulting shops, GP’s for VC and PE funds, MD's at Investment banks, and VP’s and likely even a few C Level execs at major American corporations.

For starters, the students who wrote that petition could probably put it on their resume & talk about it in an interview.

More importantly, since less than 40% of the people at a place like Ross actually bothered to learn the content, due to that strange curve that basically guaranteed the ‘top 40%’ would get A’s, you could potentially get an A by memorizing rather than learning! (not more than 10% of the questions on a given exam were designed to trip up the memorizers, and these would always be the ones the memorizers all got wrong and complained about later)

In other words, the reward system was broken--at a place like that, you didn't NEED to learn to get a strong outcome! But sigh, isn't learning the point of school?

Based on conversations with compatriots at other universities and now in corporate America, I have no reason to expect this is different elsewhere and believe it carries forward quite cleaning into many (but not all) corporate roles. These same people will send a lot of emails and be vaguely involved in many projects without actually contributing tangibly to their outcomes. Books like Deep Work and Bullshit Jobs reinforce this conclusion.

This is without mentioning the latest color of corporate delusions, like the phenomena at Zillow in which executives are confusing tokens consumed with a real measure of productivity, to the dismay of many engineers who feel that if they don't play along with the ensloppification, they'll be fired.

I really do believe that the people who create real outcomes are power law distributed and roughly follow the Pareto Principle--20% of people produce 80% of the outcome in many fields.

This is why I'm so enthralled by being close to the P&L, and learning about real tangible things.

Not everyone at Ross thrived in nonsense work. Some people saw through the bullshit. They were often:

• Studying CS or Math in conjunction with business (mote honesy feedback loop, math and code works or it doesn’t)

• Traders who were actually close to the P&L, meaning they had a brutally honest reward system

• Business owners, which I've always thought involved one of the most honest feedback loops

The commonalities between all of these people is they actually were closer to reality and the consequences of their actions.

I think that matters a lot.

[Identifying Real Work]

In most situations, there are a handful of actions you can take that will move you forward more than anything else.

For my startup, BirdDog, these actions are often related to improving the product or our 'growth engine', which includes how we market and sell the product.

There are other important things, like our finances, but these are often times a lot more bounded. I go over our books once a month for 2 hours and incidentally whenever we need to make a non trivial monetary decision.

Our bookkeeping system, though, is very emergent. Meaning I am doing the bare minimum needed to manage the finances at any given time. As an example, we did not bother tracking the states of each of our customers until the end of last year when we were making sure we didn't cross a transaction threshold for any given state.

This month, I'll be adding in another sheet to track money due to referral partners, because it's reached the point where I don't feel confident I can keep track of it in my head.

But, if I wanted to, I could have spent 10s of hours one year ago preliminarily adding these things into the finance system, along with others, like bad account provisions and AR and so many other things. It might’ve feel productive, but it would’ve been bullshit. I would’ve been solving for problems that, at the time, weren’t real. Instead, we have an ad hoc, iterative approach to building system like this, and it helps us avoid bullshit a lot.

Those same 10 hours I could’ve spent optimizing a spreadsheet were better spent getting customers or making the product better.

And even with something like product work as a category, there are things that are different levels of ROI. This weekend, I spent an hour 1.5x'ing the size of our company database. That's a way higher leverage hour than it would've been if I spent that same time debating the ideal shade of a fucking button.

You have constrained resources and you can only do so many things at once, so you really should be making sure they are the highest leverage things. Don't solve problems that don't exist.

I didn’t read the article but I must admit the concept is believable; a lot of people do create problems that don’t exist.

[You'll be Wrong]

Also to be clear, you don't have to be right about what the highest leverage thing is - you'll probably be wrong until you do it and test it and find it out.

That goes back to my post on Retardmaxxing - in more situations than we think, the cost of being wrong is lower than the cost of doing nothing. (if it’s not, it’s worth asking if there’s a way you can lower the cost of being wrong)

To be clear, though, aggressively taking action is completely congruent with not doing bullshit work; a lot of bullshit work can be identified by prioritizing talk and analysis about an important thing over doing the important thing.

When you Retardmax, you make sure the downside is actually not that deep and proceed with the action.

[Don’t Create More Work]

The most insidious kind of bullshit work is the bullshit work that you create for other people via your communication style or lack there of.

As a very subtle example, this week, I was asked by person Z if I had talked to person X who had in-bounded about Y. I said no, I hadn't.

Person Z proceeds to message me, saying great, I'll tell person X that you'll email them.

Now this might seem like a benign response from Person Z, but it actually creates work for me that simply should not exist. Now I have to:

1. Find Person X's contact info in our system of record

2. Guess at or ask Person Z what they’ve already told Person X

3. Write and send an email to Person Z that has that context in mind

None of this work is remotely necessary. The fact that it’s ‘not a lot of work’ doesn’t really matter, because it’s all a bunch of extra steps that I now have to do.

If instead of sending a message to Person X saying I would message them, Person Z would have copied me on her reply to person X saying I would take over, items 1 and 2 would be deleted, and item 3 would be easier to do.

This would have been 0 marginal work for Person Z. It is the same amount of effort as her applying to person X directly.

If it seems like I am splitting hairs over something small here, it's because these things add up. If this sort of comms style is not a big deal to you, you might be in a corporate setting that can support this sort of thing.

The difference between getting 0 of these messages a day and 5 of them can be anywhere from 10-15 additional micro items you need to do. If you don’t batch them all at once and are context switching, that really is a lot.

I value my time a TON. I, for one, don't want those additional items!

As a matter of fact, I'm deathly allergic to them.

To be clear, this isn't to say that when Jack and I communicate, this sort of thing never happens; I'm sure it does every once in a while. But, we both greatly respect our own time and each other's time and are very conscious about not wasting it to prove some convoluted point about how much work and motion we’re doing.

The person who sent me this particular message is quite the opposite, and almost always finds a way to create more bullshit work like this.

As an aside, this kind of communicator often overlaps with the person who sends an open ended message about something they're worried about. This is a classic bullshit work generator because now your job is to parse out their emotional state from real concerns and then propose a solution if appropriate. And often times, there's an implication in this sort of message that the other person wants you to solve this problem. This creates both more work and more cortisol.

Very efficient people consistently make an effort to create less work with their communications, not more.

[AI Slop Bombs]

One last, new class of communication that creates bullshit work is the AI Slop Bomb: when someone sends you a verbose message clearly written by an LLM full of action items.

These messages are very frustrating and I would argue selfish, too. It's one thing to use an LLM to summarize or recap a conversation and extract agreed upon action items, it's another thing to use an LLM to create tasks that aren't real.

I recently sent someone a proposal, and got back an LLM written response with 3 or 4 overlapping and vague clarifying questions that revealed no real comprehension of the underlying document. It was the type of response that 'looked' real but didn't mean anything.

I truly believe this was selfish and inconsiderate of the other person. I had to parse through machine generated text to try to find the actual concerns this person had. I responded, clarifying a few points and articulating what I thought he was asking for.

We got on the phone, and he read back the concerns that the LLM told him were important. We came to a working resolution, I drafted the changes and sent back a new proposal. Once again, he responded with an LLM. One of the questions was literally asking for clarification about a definition that an entire fucking exhibit was dedicated to.

By not actually just reading and processing the document, or validating what the LLM was saying, this person is creating more work for me that I don't want to & shouldn't have to do. I'll send a very short response to each point, but would be lying if I said I didn't consider volleying AI slop back.

LLMs amplify our ability to create bullshit work, which is a very unfortunate thing for all of us.

I don’t even known what I write about anymore, but subscribe if you like it - I’m here each Sunday.

[Do the Hard Thing]

All hope is not lost.

If anything, from a competitive standpoint, the amount of people doing bullshit work should make it easier for you to get an out-sized return by doing real work. And maybe, if enough of us do that, we'll turn the tide against nonsense.

Remember, this real work is inherently challenging by definition.

On one end, it might be the deep work that is harder to do than shallow work. Hence, less people do it, and the return for doing it is higher.

On the other end, it might be that the real work is a certain kind of uncomfortable situation you've been avoiding (a weak spot, if you will). If you've been avoiding this thing, then it's likely that it's also where the comparatively 'easy wins' are hiding.

A weak point of mine is I don't approach random people in public. Heard three guys talking about a startup they were building in the park the other day, but didn't do anything. They could be future guests at one of my hacker house's events or even residents.

So, I told a cop in the coffee shop thank you for his service while writing this. Small start. Find the weak spot and break it, or the muscle that atrophied and grow it.

The more you focus on identifying and doing the high leverage thing, the cooler the outcomes you'll get. With BirdDog, as I've been overcoming my fear of building a product that users can use without talking to us, we've already been seeing some very attractive outcomes.

It's been hard product work, but I know right now this & articulating these changes to the market are more important than anything else for the outcome we want.

Live Deeply,