Valley Recap

🪙 Nvidia’s Backstop Universe – Heads I Win, Tails Who Loses? 🌩️NEOCLOUD SUMMIT OCT8 💰 Bay Area Startups Collectively Secured $10B+ in September MTD

Sat, 12 Sep 2026 15:00:55 +0000

Nvidia’s Backstop Universe – Heads I Win, Tails Who Loses?

(Semianalysis Article by Daniel Nishball, Oliver Kennon, and Terence Ong)

The $11T AI Buildout, Nvidia’s Backstop Economics, and the Limits of Nvidia’s Balance Sheet

An expanded view of Nvidia’s Balance Sheet

As of 2Q F1/27 (second quarter of fiscal year ended Jan 2027) Nvidia in its 10-Q disclosed $530B of gross off-balance sheet guarantees across six different line items – a jump from $184B of obligations disclosed in the prior quarter due mainly to moves in three items:

Supply and capacity commitments went from $119B to $279B, primarily memory as explained by the CFO, with 96% of these supply commitments due by F1/29.
Guarantees and LPS Guarantees went from $3.5B to $108.5B on SB Energy’s PORTS-Pike campus in Ohio, representing 4.25 GW leased to OpenAI for twenty years with Nvidia guaranteeing the land, power and shell.
Two line items appeared for the first time: $36B of AI cloud agreements, the take-or-pay floors under Neocloud capacity, and $20B of datacenter leases Nvidia has signed as a tenant but expects to reassign to third parties.

The $530B of commitments dwarf on-balance sheet liabilities of $91B, which includes $33.4B of total debt with $25B of senior notes issued in the most recent quarter. On the asset side, Nvidia’s cash balance has increased from $12B to $22B in the past few quarters, but its investment book (marketable and non-marketable securities) has leapt from $45B in 2Q F1/26 to $128B by 2Q F1/27. The $99B equity book generated $23.7B of gains in the first half, and $25B more of equity investments are committed.

The question on everyone’s mind is this: given Nvidia’s on-balance sheet and off-balance sheet obligations, what is its capacity to meet all of these commitments in the case of an industry slowdown or downturn, and how much can Nvidia grow these obligations over time while still enjoying a margin of safety? How much support does the industry need and how much support should Nvidia offer?

In the chart below, we illustrate the history of these off-balance sheet items as well as our estimates of how much additional obligations Nvidia can reasonably support given its balance sheet and future cash flows. To be clear, this does not represent our base case or a definitive forecast for where we think Nvidia will go with its various backstops. Rather it focuses on the extent of Nvidia’s ability to support obligations given its business profile, and we argue given this considerable firepower, that it should have the willingness to continue expanding these commitments significantly.

Source: SemiAnalysis AI Compute, Capital and Markets Model

What is obvious from the diagram above is that Nvidia is a true behemoth when it comes to cash generation – consensus estimates have it generating ~$441B of EBITDA in F1/28, and this is why we reflect this via a growth of the cash balance to $1.4T by F1/31 – though in reality Nvidia will most likely reinvest this cash into various securities. These are huge numbers, yet the scale of overall AI investment is larger – we estimate ~$11T of cumulative capex from CY24 through CY29!. This note will explore how far Nvidia can take each form of support, and what impact this could make on the massive AI investment needs ahead of us.

SemiAnalysis AI Compute, Capital and Markets Model

Why all the Guarantees and Backstops?

The mechanism that gives rise to these off-balance sheet guarantees and backstops is the same one we described in The Front End Gets Crowded (institutional article). The Gigascalers (Amazon, Microsoft, Google, Meta, Oracle) have been gatekeeping the investment grade capital that the AI buildout runs on and have benefitted tremendously from the high spread between offtake price they are being charged, and spot/short-term price that they are selling at as well as services that sit on top. In response, Nvidia has been manufacturing alternative credit anchors so Neoclouds and Neolabs can raise money without the Gigascalers. In these support arrangements, Nvidia sells GPUs at full margin on day one, then floors the buyer’s revenue (in the case of the AI Cloud Partner (AICP) Program), or guarantees the buyer’s landlord, or signs the lease itself, all so that non IG-rated operators can effectively borrow at IG rated pricing.

Heads, demand holds and Nvidia wins twice, once on the GPU sale at full margin and again on the revenue share above the floor. Tails, demand falls and the Neocloud makes no money but remains solvent, as the revenue floor is set high enough to repay lenders. Nvidia’s backstops and guarantees also act as a breakwater against further contagion as lenders have some shielding from a downturn. Yet how high and broad the breakwater that is the Nvidia backstop/guarantee depends on Nvidia’s ability to meet these obligations.

Risk reward is asymmetric in Nvidia’s favor as long as its balance sheet remains strong. Nvidia loses only if the backstopped Neoclouds fail to meet their lease and offtake commitments at the same time that Nvidia’s own business and cash generation slow, degrading its ability to make good on these obligations. The two are correlated, since a Neocloud that cannot pay its lease is a Neocloud that has stopped buying GPUs. Only if a downcycle is steep enough to overwhelm Nvidia’s ability to meet its off-balance sheet obligations, then it will be tails, we all lose.

By our count, Nvidia today backstops ~6.5 GW of capacity, most of which is not yet built. Our datacenter model has Microsoft, Meta, AWS and Oracle leasing roughly 15 GW of third-party capacity in 2026 and we expect them to lease more than 35 GW by 2028. Each of those leases typically run fifteen to twenty years, and developers borrow against them with investment grade pricing. The Gigascalers are the implicit backstop of the buildout and are in aggregate far larger force than Nvidia’s backstopped capacity, yet Nvidia’s program gets the attention because it is new and because many investors have Cisco PTSD, contracted vicariously without an appreciation for the dissimilarities.

Source: SemiAnalysis AI Compute, Capital and Markets Model

We will now explore how this support could evolve in detail – walking the six off-balance sheet obligation lines in the recent quarter’s disclosure, plus a new one we have created to track residual value guarantees under the capital partnership.

AI Cloud Agreements

This line item made its debut in the 2Q F1/27 earnings report released on August 26th, 2026 at $36B of aggregate obligation, spread across six years. It represents Nvidia’s AI Cloud Partner (AICP) program, where Nvidia backstops GPU rental contracts for six years at a below market rate, approximately $2.35/hr/GPU on average for the GB300 vs the $4.50 to $4.60 range for 5y contracts, with the intent of making these AI clusters bankable. Neoclouds in this program would then seek to rent out compute on shorter term contracts to AI Natives and other tenants, but can always rent to Nvidia at any time should these contracts fall through or not renew. We published an extensive analysis of this structure in a July 2026 Newsletter Article.

Source: SemiAnalysis AI Compute, Capital and Markets Model

There are three deals announced so far under AICP: Firmus’s 360MW deal in Batam corresponds to a $21.1B total obligation, the SharonAI deal means $4.2B of obligations and we estimate the GMI deal corresponds to $2.2B of obligations for Nvidia. We believe there are still a few deals that have been completed before Nvidia reportedly paused AICP two weeks ago and estimate this line item will close F1/27 at $51B.

We do understand that Nvidia is not advancing any further AICP deals at this time, but believe that they still wish to be supportive in one form or another. In this analysis, our forward modeling still assumes considerable growth in this line item to reflect our view that Nvidia should offer support in some form, as well as to explore how far Nvidia could go in terms of explicit support.

Source: SemiAnalysis AI Compute, Capital and Markets Model

Residual Value Guarantees under Capital Partnerships

Earlier in August 2026, Nvidia announced its partnership with a number of private equity funds to mobilize over $500B of third-party capital to support the buildout of AI Clusters. Under this program, institutional investors are in the driver’s seat and will evaluate and price each proposed project, with the only support from Nvidia being a residual value guarantee for up to 25% of a deal. In his X post detailing the program, Jensen Huang highlighted that this support would be limited and on a project-by-project basis. Though there are few details on what such a program could look like, we drew inspiration from the Google-Broadcom-Anthropic-Apollo TPU structure to surmise how deals could be structured.

In the TPU backstop structure, if Anthropic is unable to continue to service lease payments to the TPU SPV, then the structure can be resolved by Broadcom assuming the lease or by the racks being sold. If the racks are sold, then the residual value guarantee takes effect, as Broadcom must then make the senior noteholders whole after such equipment sale – effectively plugging the residual value gap, if any. The deal gets better – Google will play its part by also backstopping rental payments to the various datacenter operators. The effect is that lenders to the TPU SPV are effectively taking Broadcom credit risk while lenders to the datacenters are taking Google credit risk.

Nvidia’s proposed structure would appear to wean investors away from this dual backstop as it not only reduces the residual value guarantee to only up to 25%, but it also is silent on any datacenter backstop.

Source: SemiAnalysis AI Compute, Capital and Markets Model

We think that the funds could tranche the lending to create different pools of risk, with the first loss being taken by riskier equity/mezzanine style lending and the last loss taken by more senior creditors. Nvidia’s 25% residual value guarantee would insert itself above the senior lenders but after the equity/mezzanine investors, adding a meaningful amount of crumple room to the structure, which could place the senior tranche at low enough risk of actual shortfall to attract conservative lenders like banks.

This structure can also provide the most bang for buck when it comes to Nvidia’s off-balance sheet obligations. While the AICP creates $59B of obligations per GW of capacity enabled and the PORTS-Pike structure requires $25B of obligations per GW, the residual value structure would only require $9.4B of obligations per GW enabled, though the datacenter leg still needs to be solved for.

While Nvidia has said the program targets over $500B of capital, we think it will likely swell larger due to the strong compute demand. In the spirit of examining the full extent of Nvidia’s maneuverability, we take an aspirational stance, assuming $2.5T of total funding under this program through F1/31, with the obligations under this line item ending F1/27 at $55B before growing to $373B by F1/31.

Source: SemiAnalysis AI Compute, Capital and Markets Model

Tickets Are Limited

Check out our video library of over 300+panels, presentations and interviews

JOIN US AT THE MIDWAY!

Bay Area Startups Collectively Secured $10B+ in September MTD

Funding season is underway – ten megadeals in four days, more than $10B in deals in September already. Cognition AI, Positron AI, Kepler Computing, Encoded Therapeutics, Upwind Security and Harvey were among those fundings, details on each of those companies below. That $10B takes the YTD startup funding total over $375B, and the current run-rate would end the year at over $500B.

Startups: Looking for partners, keeping a watch for possible competitors? LinkSV's Pulse of the Valley weekday newsletter and alerts can keep you on top of new companies launching; database keyword search gives you prospects for partners and customers. Founders get the newsletter, and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to get early notice of new tech, stay on top of SV funding intelligence, and keep track of the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Maven Robotics closed a $100M Series A, building the world’s leading general-purpose AI robots.
Poseidon Aerospace closed a $60M Series A, build and operate full-size autonomous cargo aircraft to move goods.
Cradlewise closed a $12M Series A, a smart sleep company with a smart all-in-one convertible crib and AI that learns your baby’s unique sleep patterns.
OpenRelay Incorporated closed a $9M Seed, a distributed GPU cloud for production inference, hardware-agnostic, token SLA-backed inference up to 90% cheaper than AWS, Azure, and GCP.
Tellia Inc closed a $5M Pre-Seed, developing Voice OS for the field and reinventing how field operations capture and use data starting with agriculture.

Growth Stage:

Cognition AI closed a $2B Series E, an applied AI lab building end-to-end software agents.
Positron AI closed a $875M Series C, delivers faster inference, lower power usage and lower total cost of ownership for both enterprises and research teams,
Upwind Security closed a $300M Series C, the next-generation cloud security platform built to lead the Runtime revolution.
Encoded Therapeutics closed a $275M Series F, creating one-time, disease-modifying gene therapies for pediatric central nervous system (CNS) disorders
Kepler Computing received a $245M Grant, a new frontier of AI memory that approaches the bandwidth per watt of SRAM and goes up to 10x beyond the capacity of SRAM and HBM.

Your Feedback Matters!

Your feedback is crucial in helping us refine our content and maintain the newsletter's value for you and your fellow readers. We welcome your suggestions on how we can improve our offering. engage@ignitegtm.com

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

☁ NEOCLOUD SUMMIT: Where the Next Cloud gets BUILT 💰 Bay Area Startups Collectively Secured $3.7B in September Week 1 🇺🇸 Happy Labor Day

Sat, 05 Sep 2026 18:33:37 +0000

Where the Next Cloud Gets Built

Neoclouds sit at the center of the AI infrastructure universe. Where our AI Infra Summit (DEC3) takes the wide lens across providers, developers, and enterprise buyers. NeoCloud Summit does something different. It is a focused conversation about neocloud systems and everything that touches them: the layers below, above, and around the operators building AI-native capacity.

The Ecosystem in one room. 🔊SOUND ON 🔊

This is where the next cloud gets built. And right now, nobody has it fully figured out.

Supply chains are fragmented. Partnerships are still forming and reforming. The economics are being priced faster than anyone fully understands them. This is a frontier ecosystem, and frontier ecosystems do not get solved by one company in a room alone. They get worked out when the people building them compare notes in the same place at the same time. That is why we bring this together.

It is also why we build the program the way we do. We do not curate a lineup a year in advance and fill a stage with logos. We build around what is actually salient right now, so the conversations on stage match the ones the market is having this quarter. On a topic moving this fast, relevance is the whole point.

Here is some of what the day digs into:

What is actually changing. The constraint has moved beyond GPU availability. Cooling, land, financing, supply chain, and deployment timing now shape outcomes as much as chips do. The market is evolving from H100-era buildouts toward Rubin-era infrastructure, and the pace of that change is forcing new decisions across the entire stack.

The financing question. Everyone is raising and building. Fewer are talking openly about the unit economics underneath it: GPU depreciation, the capital stack, and what holds up if utilization or pricing softens. Tied to it is utilization itself, the difference between having GPUs and having GPUs that actually earn. This is one of the least-discussed and most important conversations in the space.

Why neoclouds have to be more than bare metal. If everyone is renting the same GPUs, what keeps neoclouds from becoming interchangeable? The strongest operators build a full stack and differentiate on reliability, service design, and knowing the customer workload, not simply on having capacity.

The shift to inference. The money is moving from training to inference, and inference has completely different infrastructure requirements: latency, locality, cost per token. Neoclouds optimized for training buildouts may be pointed at the wrong workload, and that gap is opening in real time.

What enterprises actually want. The buyers close the day. The real reason enterprise AI stalls is not compute. It is trust: security, reliability, and integration into existing environments that are already messy. Buyers do not want capacity. They want outcomes they can run in production.

The throughline is simple. The winners over the next 6 to 12 months will pair capital and GPUs with differentiated operations, targeted products, and infrastructure people trust. Everyone else will be counting GPUs.

If you build, buy, fund, or power neoclouds, this is the most relevant AI infrastructure conversation you can be in this year. That is what we are building on October 8.

NeoCloud Summit 2026 — Where the Next Cloud Gets Built · Luma

The neocloud market is being built right now, and the people building it will be in one room on October 8. NeoCloud Summit brings together the full stack of…

luma.com/neocloud_summit_oct8

Check out our video library of over 300+panels, presentations and interviews

Tickets Are Limited

JOIN US AT THE MIDWAY!

Bay Area Startups Collectively Secured $3.7B in September Week 1

It was a slow week, the lull before Labor Day - but the deals total still broke the $3B mark in an AI-heavy week where five megadeals were 93% of fundings.

If there was any doubt that open-source models will take an increasing percentage of the AI workload, NVIDIA's $13B acquisition of Hugging Face eliminates it. In response to the frontier labs and Big Tech all working towards less dependence on NVIDIA by building their own chips (Anthropic was the last holdout), NVIDIA will invest and ensure that open-source models are a competitive alternative.

Most August funding ever, why and to who? New technology in silicon, power and compute, what to expect as we end Q3 and into Q4. Check us out next Friday, register here.

Startups: Looking for partners or possible competitors? LinkSV's Pulse of the Valley weekday newsletter and alerts can keep you on top of new companies launching; database keyword search gives you prospects for partners and customers. Founders get the newsletter, and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to get early notice of new tech, stay on top of SV funding intelligence, and keep track of the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Emperik.ai closed a $21M Seed, building the autonomous infrastructure engineer.
Andromeda Surgical closed a $15.1M Series A, developing an autonomous surgery platform.
Visko Platform closed a $10M Pre-Seed, an AI research company pioneering Live Models — a new class of world models that run rather than render.
Robocurve closed a $.5M Pre-Seed, developing open-source evaluation frameworks and independent benchmarking platforms for robots and the AI models that control them.

Growth Stage:

Crusoe closed a $3B Series E, provides a reliable, scalable, cost-effective, and environmentally friendly solution for AI infrastructure.
Gimlet Labs closed a $300M Series B, combines GPUs, near-memory compute, dataflow architectures, and CPUs in one heterogeneous system for 10X gains in throughput.
Lyte AI closed a $165M Series C, designs custom silicon, multimodal sensors, spatial software that give robots and autonomous systems a real-time view of the physical world.
Greptile closed a $40M Series B, an AI expert that understands your codebase, as an API.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Blu Dot surpasses 2,000% ROAS with self-serve CTV ads

Home furniture brand Blu Dot blew up on CTV with help from Roku Ads Manager. Here’s how:

After a test campaign reached 211,000 households and achieved 1,010% ROAS, the brand went all in to promote its annual sales event. It removed age and income constraints to expand reach and shifted budget to custom audiences and retargeting, where intent was strongest.

The results speak for themselves. As Blu Dot increased their investment by 10x, ROAS jumped to 2,308% and more page-view conversions surpassed 50,000.

“For CTV campaigns, Roku has been a top performer,” said Claire Folkestad, Paid Media Strategist, Blu Dot. “Comping to our other platforms, we have seen really strong ROAS… and highly efficient CPMs, lower than any other CTV partner we've worked with.”

Using Roku Ads Manager, the campaign moved from a pilot to a permanent performance engine for the brand.

Learn More

THE PILOT IN THE MACHINE💰 Bay Area Startups Collectively Secured $17.7B+ in August MTD

Sat, 29 Aug 2026 15:00:55 +0000

We already made the case that AI strategy is really an org design problem. That argument stands. The next question is the one we actually argue about in the field, between meetings, on the drive home from a conference: what are the teams winning with AI doing differently, day to day?

We keep having a version of the same conversation. Smart people point out that AI leads to cognitive decay. Everybody just looks for the answer, the tool hands it over, and the thinking muscle goes soft. It is a sharp point, and the risk is real. But our answer is that it is not a but, it is an and. There are two paths, and the one you end up on is a choice you make in how you build.

Path A is the one the skeptics are pointing at. The tool is a black box. You ask, it answers, you copy, you move on. The person sits at the end of the workflow, not inside it. Nobody learns anything, because there was no thinking in it. Do that a thousand times and the muscle atrophies. If that is how you deploy AI, the decay is real.

Path B is the one we are living in at Ignite, and it produces the opposite effect. The pilot sits in the middle of the workflow. There are very few black boxes inside our org. The tool is a partner they reason with, not a vending machine they pull from. When the machine gets better, the pilot gets stretched, not sidelined. They take on harder problems, ask sharper questions, and stay more engaged. The neurons fire faster, not slower. Instead of cognitive decay, we get cognitive momentum, and it compounds every day we fly.

Same technology, two outcomes. The difference is not the model. It is whether you designed the human into the center of the work or out of it.

That is why the pattern is counterintuitive. The more capable the tools get, the more human skills decide the outcome. Judgment, first: AI will generate ten plausible answers in seconds, but knowing which one is right for this customer, this deal, this moment is still a human call. Empathy, second: when a buyer is nervous about handing part of their business to software, no model closes that gap, a person does. And cross-functional translation, third: AI makes it cheap to produce work inside any single function, but someone still has to sit between the technical reality and the customer promise and make sure they match.

We think about the tools like an Iron Man suit. The suit is incredible, but worthless without the person inside it making the calls. A new hire at Ignite test-flies the suit in week one and ships real work faster than anyone could a few years ago. What we hire and coach for is the pilot. That is Path B in practice. Most companies are building better machines. We are building better pilots, and the two evolve with each other.

This is not a framework we sketched on a whiteboard. It is what happens because we are flying every day. Every deal, every customer call, every internal build is another rep in the suit. This is really just us describing what the flight logs already show.

Conway's Law is worth remembering. Organizations ship their own structure, and the tools amplify whatever is already there. Path A gets faster, and so does Path B. The tool does not pick for you.

So the takeaway for anyone building a team right now is not to chase the longest list of AI tools. It is to ask which human skills your people need to get more out of the tools they already have. Hire for judgment. Reward the person who reads the room. Protect the ones who translate between engineering and the customer.

The more AI we get, the more human we need to be. Not as a slogan. As a hiring plan, and as the way we fly every single day.

Check out our video library of over 300+panels, presentations and interviews

Tickets Are Limited

Bay Area Startups Collectively Secured $17.7B+ in August MTD

Total dollars fell but the number of deals went up this week as we passed $17B with one business day to go. This week's six megadeals made up just 63% of the total, down from 93% for last week. The week included multiple early stage and some late. More AI infrastructure week's fundings included multiple startups focused on AI infrastructure startups, including Emerald AI (connecting grid and AI infrastructure) and Agentrys (agentic AI for semiconductor design).

Browse the full list here:

AI Infrastructure companies August week 4.

Startups: Looking for partners or possible competitors? LinkSV's Pulse of the Valley weekday newsletter keeps you on top of new companies coming out of accelerators and a quick search of the database by keyword gives you targets to investigate and approach. Founders get the newsletter, database and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to get early notice of new tech, stay on top of SV funding intelligence, and keep track of the companies, investors and executives impacting the startup ecosystem..

Early Stage:

Emerald AI closed a $150M Series A, provides an intelligent interface between power grids and AI infrastructure.
Deep Cogito closed a $43M Series A, builds frontier open-weight models and creates pecialized models trained on enterprises' proprietary data, decisions and outcomes.
Voya Energy closed a $35M Series A, converting abundant, high-energy, recycled metals into a powerful, safe, durable fuel.
Agentrys closed a $19.1M Seed, enabling semiconductor design teams to build, own, and continuously improve their own agentic engineering workforce.
OneTriangle closed a $.5M Pre-Seed, building the cheapest, fastest inference, cutting inference prefill costs by 20% and time-to-first-token by 40%.

Growth Stage:

Generalist AI closed a $200M Series B, an AI robotics company building general intelligence for the physical world and making it useful to everyone.
Gatik closed a $200M Series D, autonomous Middle Mile logistics, delivers goods safely and efficiently using its fleet of light and medium duty trucks.
Liner closed a $36.1M Series C, an evidence-first AI research platform designed for people who need answers they can verify.
Lilac Solutions closed a $34.4M Series D, their ion change technology enables high lithium recovery with minimal water use and no evaporation ponds.
Celera Semiconductor closed a $30M Series B, redefining the future of analog IC design.

Tensor Machines

AI systems increasingly depend on information coming directly from the physical world. Sensors, GPUs, circuit boards, autonomous systems, and other connected hardware are producing the data these systems use to make decisions.

That creates a fundamental problem. How do you know the hardware is authentic and the data coming from it can actually be trusted?

For operators, the need is straightforward: continuously verify the physical systems underneath AI, rather than assuming they are behaving as expected.

That is where Tensor Machines fits.

Who they are
Tensor Machines is building a physics based hardware authentication and sensor data integrity platform for AI infrastructure and cyber physical systems.

What they deliver
Their technology uses the physical signatures naturally produced by electronic hardware, including electromagnetic, thermal, and power signals, to create a unique fingerprint of a device. Tensor Machines can then use those fingerprints to continuously verify hardware authenticity and integrity while equipment is operating.

That gives operators another layer of visibility for identifying altered or counterfeit hardware, detecting physical layer anomalies, and validating the integrity of sensor data feeding critical systems.

Who they serve
The technology is relevant across AI infrastructure, autonomous systems, robotics, defense, industrial environments, and other applications where connected electronics and sensor data have to remain trustworthy.

Tensor Machines was also selected as a 2025 DataTribe Challenge finalist, bringing its approach to physical layer security in front of cybersecurity leaders, investors, and potential customers.

As AI gains greater control over physical systems, the integrity of the hardware and sensor data underneath those decisions becomes increasingly important.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Blu Dot surpasses 2,000% ROAS with self-serve CTV ads

Blu Dot used Roku Ads Manager to drive incredible results for its furniture sales event. Its strategy hinged on custom audiences and retargeting, where intent was strongest.

“Roku has been a top performer,” said Blu Dot’s Claire Folkestad. “We have seen…CPMs lower than any other CTV partner we've worked with.”

Learn More

YOUR AI INFRA CALENDAR FOR BACKHALF 2026 💰 Bay Area Startups Collectively Secured $16B+ in August MTD

Sat, 22 Aug 2026 15:00:55 +0000

Six months ago we started planning the busiest stretch IgniteGTM has ever run. Six events between September and December, each one chosen to build on the last.

Here's how they connect.

It starts September 17 in South San Francisco with AI for Life Sciences. Biotech puts real weight on AI infrastructure. Genomic pipelines, drug discovery, and clinical data that can't leak and can't wait. The requirements here are specific: different workloads, different security constraints, different performance demands than general enterprise AI. Starting here sets up a question the rest of the fall answers: how infrastructure changes to fit the workload.

Request To join

A week later, on September 24, we're in New York for the AI for Financial Services Summit. Financial services runs into its own version of the problem: latency that costs money, compliance that can't bend, and data that lives under heavy regulation. Same theme as Life Sciences, different constraints, two industries showing why one-size-fits-all infrastructure doesn't hold up.

Join Us in NYC

Next is our first-ever NeoCloud Summit in San Francisco on October 8. Neoclouds are building GPU-dense, AI-native infrastructure as an alternative to the hyperscalers. Over the past two years that movement has grown in pieces: individual raises, new data centers, founders with a thesis. The Summit puts those founders and operators in one room to trade notes on where the industry goes next.

Grab Your Ticket Today

On October 13 the OCP Networking Party is on our home turf in San Jose, the middle and the heart of Silicon Valley, with Supermicro and AMD, sitting alongside OCP Global Summit and the nearly 11,000 people building the open hardware stack. Then we take it on the road: SuperCompute 26 in Chicago on November 17, the year's biggest HPC and AI gathering (SC24 drew more than 18,000), where the hardware teams, hyperscalers, researchers, and operators are all in one place. Both parties sit inside those rooms, so we tap the full crowd of each conference.

All of it feeds AI INFRA SUMMIT 6 on December 3 in San Francisco. What we learn from September to December, what life sciences needs, what financial services demands, what the neoclouds are betting on, and what the hardware ecosystem is shipping, goes into the AIS6 agenda. It's the flagship, on a bigger stage, pulling the full infrastructure stack into one room.

Early Bird Tickets On Sale

We close the year with Enterprise AI Day on December 16 at Santana Row in San Jose. This one measures whether the rest of it is working in production. Leaders running real AI at scale, with real budgets and real consequences. Life Sciences and Financial Services open the year's questions, AIS6 puts them on the main stage, and Enterprise AI Day checks the results inside the companies that have to make the economics work.

The sequence tracks how the industry is actually moving: specialized workloads (Life Sciences, Financial Services), the infrastructure layer built to serve them (NeoCloud), the broader ecosystem shipping it (OCP, SuperCompute), the full stack convening (AIS6), and the enterprises putting it into production (Enterprise AI Day). Every venue and speaker was picked to move that sequence forward.

We've been heads down building this for months. It's the most ambitious run we've attempted, and we're just getting started.

Whether you're building this infrastructure, buying it, or investing in it, there's a room for you.

See you out there.

Check out our video library of over 300+panels, presentations and interviews

Get the MasterPass Today

Tickets Are Limited

Bay Area Startups Collectively Secured $16B+ in August MTD

It doesn't feel like summer - funding activity is not slowing down. There were eleven megadeals for $4.9B, 93% of the week's $5.3B total The week's megadeals included multiple AI infrastructure startups – Groq (neocloud), Etched (high speed inferencing) and Velaura AI (fka Auradine – pivoting from building high-efficiency Bitcoin mining hardware to ultra low power AI compute).

Early this year, Groq was the target of a reverse acquihire as NVIDIA hired their CEO and top talent and licensed Groq's technology, leaving the company to re-invent itself. Groq pivoted to being a neocloud provider using NVIDIA hardware, and has $1B to begin scaling from 54 to 200+ megawatts to support inferencing.

For startups raising capital: Drowning in lists? Find investors in companies like yours with LinkSV. Stay on top of who's raising, who's closing and who's investing with the Pulse of the Valley weekday newsletter. Founders get the newsletter, database and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to get early notice of new tech, stay on top of SV funding intelligence, and keep track of the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Groq closed a $350M Series A, operates the inference cloud infrastructure that runs real-time AI.
Velaura AI (fka Auradine) closed a $110M Series A, develops ultra-low-power compute technologies that enable the next generation of AI infrastructure.
Thunder Compute closed a $13M Series A, a virtualized GPU cloud platform for building AI/ML.
idler closed a $9M Seed, building the evals and environments that frontier labs use to measure and train their models.
Synthefy inc closed a $6.5M Seed, building multi-modal generative AI models for time series data.

Growth Stage:

Etched closed a $700M Series D, builds frontier inference clusters designed to make AI inference dramatically faster and cheaper.
Higgsfield AI closed a $400M Series B, an AI-native multimedia content creation platform for creators, marketers, brands, agencies, and studios.
Wispr Flow closed a $280M Series B, smart voice-to-text that helps you work 4x faster in any app.
Muon Space closed a $250M Series C, designing, building, and operating high-performance satellite constellations for defense, civil, and commercial customers.
Fuse Energy Technologies closed a $100M Series B, a nuclear fusion company dedicated to accelerating the world's transition to fusion energy.

AI applications are getting more demanding. Reasoning models need to generate longer outputs, agents are chaining together multiple calls, and interactive applications cannot afford to leave users waiting while inference catches up.

For builders, inference speed is starting to influence what products can actually be built and how people interact with them.

That was the focus at Cerebras Supernova.

What it is
Supernova is Cerebras’ flagship event, bringing together developers, founders, researchers, enterprise AI leaders, and infrastructure partners working on the next generation of AI applications and compute.

Why this, why now?
The event put fast inference into practical context. Cerebras showcased live production AI applications alongside technical talks, customer deployments, product updates, and a CEO keynote from Andrew Feldman.

The conversation matters because faster inference changes the experience at the application layer. Reasoning can happen closer to real time. Agents can complete more steps without creating long waits. Developers can start designing around responsiveness that previously was difficult to achieve at scale.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Nobody actually knows how AI gets used.

AI tool sprawl happened fast, and visibility never caught up. Employees paste contracts into ChatGPT, run code through Copilot, and build workflows in apps security never approved.

Harmonic Security classifies every AI interaction by task, tool, and team, so you can see which use cases drive real productivity, which tools are shelfware, and where sensitive data is headed. Across approved and unapproved apps alike.

No need to guess what 'AI adoption' means inside your company.

Explore Your AI Usage

Key Takeaways from Building the AI-Native Startup Stack 💰 Bay Area Startups Collectively Secured $10.7B+ in August MTD

Sat, 15 Aug 2026 15:00:55 +0000

On August 14, roughly 200 founders, engineers, investors, and AI builders gathered at The Canopy in Menlo Park for "Building the AI-Native Startup Stack" a full-day program of keynotes, panels, live demos, and networking hosted by Ignite, Startup Grind, and Tencent Cloud.

The Next Advantage Isn't Model Access

If there was one theme that echoed across every session, it was this: the companies winning with AI aren't the ones with the best model. They're the ones with the cleanest data, the sharpest workflows, and the discipline to operationalize intelligently.

Anderson Lin, General Manager of Tencent Cloud, put it directly in his keynote: the industry has moved fast through infrastructure and models, and is now squarely in the agent layer where execution matters more than experimentation. The biggest blocker he sees in enterprise AI adoption is messy internal processes and disorganized knowledge bases.

Inside Tencent's Own AI Transformation

One of the most striking reveals: over 90% of Tencent's code is now written by LLMs using an internal tool called Cobalt. WeChat, with 1.6 billion daily active users has integrated agentic tools. Two new AI-powered sales tools launched internally the same day as the event. Tencent is restructuring core workflows around AI across engineering, sales, communication, and support.

Voice AI Has Graduated

The voice AI panel moved past "does it sound human?" into something more useful: does it behave like a high-performing human in that specific role? Latency is no longer the headline challenge sub-second response times are common. The harder problems now are judgment, memory handoff, context continuity, and accuracy on things like phone numbers and account IDs where one wrong character breaks the experience.

A standout example: one panelist described calling a hotel, telling his story to an AI, getting transferred to another AI and repeating it, then getting handed to a human who also asked him to retell it because no one had transcript access.

AI Media Is Shifting from "Generate" to "Direct"

The media panel tracked an evolution from generation (wow-factor clips) to production (repeatable, consistent output) to simulation (interactive worlds users can shape). AI has pushed the first 80% of creative work close to free, but the final 20%; taste, vision, and judgment still requires a human hand. A new role is emerging: the "AI Director," a creator who orchestrates the full workflow from concept to publish, often solo.

"Prompting is not creativity."

Token Economics Are a CFO Conversation Now

Cheaper tokens haven't lowered total spend. They've increased it more experimentation, more agents, more usage. Organizations are moving toward per-application token tracking, smart routing (stronger models for planning, cheaper models for simple tasks), and governance frameworks that prevent agent sprawl. One panelist contrasted a three-day engineering task with a 15-minute agent run costing $7 in tokens compelling ROI, but only if you're tracking it.

The Bottom Line

AI infrastructure has matured past the novelty phase. The frontier now is operational who can structure data, design agentic workflows, control costs, and deploy reliably at scale. As I put it in my opening remarks: "The more AI we get, the more human we need to be."

Check out our video library of over 300+panels, presentations and interviews

Get the MasterPass Today

Tickets Are Limited

Why This Audience and This Place Matters

Silicon Valley is still a collection of microcultures, each with its own role in the broader innovation economy. San Francisco often represents the software conversation, San Jose reflects the hardware and infrastructure layer, but Menlo Park and Palo Alto remain the center of gravity for startups, venture capital, and the earliest stages of company building.

That’s what made this audience special. The room brought together founders actively building, investors backing the next generation, and ecosystem partners helping turn ideas into companies. In a region defined by innovation density, Menlo Park still stands out as the place where capital, ambition, and company formation collide most directly. And that was the point of hosting the event here: not just to gather people in Silicon Valley, but to bring them into one of the Valley’s most important startup corridors, where conversations turn quickly into partnerships, funding, and momentum.

Bay Area Startups Collectively Secured $10.7B+ in August MTD

August funding activity spiked this week, with nine megadeals totaling $8.1B, 93% of the week's $8.7B. The week's deals included multiple AI infrastructure startups – River AI, Pathway, Trajectory and Core Automation on the LLM/AI Research lab side; Source Foundry, Point2 Technology, Oxide Computer Company and MicroRefrigeration on the hardware/chip/cooling side. More on each below.

2026 VC startup funding YTD has already exceeded 2025's full year total, breaking the $350B mark in July. If deal numbers and dollars stay at the same level, we're on target to break $500B by the end of the year.

Follow LinkSV on LinkedIn to get early notice of new tech, stay on top of SV funding intelligence, and keep track of the companies, investors and executives impacting the startup ecosystem.

Early Stage:

River AI closed a $1.1B Series A, develops software and APIs for building personal AI systems that users can shape and control.
Source Foundry closed a $400M Series A, aiming to revolutionize microchip manufacturing and reduce production costs.
Trajectory closed a $40M Series A, the platform for continual learning, turning real product usage into AI that continuously improves.
Pathway closed a $15.5M Seed, introducing the world's first post-transformer model that adapts and thinks just like humans.
MicroRefrigeration closed a $5.5M Seed, develops next-generation AI infrastructure cooling solutions centered on a proprietary vapor-compression technology.

Growth Stage:

Databricks closed a $5B strategic round, is the Data and AI company.
Oxide Computer Company closed a $445M Series D, a new kind of server that is true rack-scale design and built with the innovations of cloud hyperscale technology.
Core Automation closed a $432.1M Series B, building the world's most automated AI lab.
Point2 Technology closed a $136M Series B, develops ultra-low-power, ultra-low-latency RF and mixed-signal interconnect SoCs for AI and accelerated computing.
Skan closed a $63M Series C, provides a continuously evolving record of how work actually happens across every system, application, and exception for enterprise AI.

Axiado

AI servers are getting denser, more expensive, and more complex. At the same time, the systems responsible for controlling and managing that hardware have privileged access to the infrastructure underneath the workload. A compromise at that level can sit below many of the security tools organizations traditionally rely on.

For operators building large AI clusters, the outcome they need is straightforward: establish trust at the hardware level, detect abnormal behavior early, and contain threats before they spread across critical infrastructure.

That is the problem Axiado is tackling.

Who they are
Axiado develops hardware anchored platform security and system management technology for AI data centers, cloud infrastructure, networking, and 5G.

Ben Levine, Sr. Dir., Product Mgmt. & Marketing, Axiado with Linda Yang of Supermicro in our Ignite Studio

What they deliver
Axiado's Trusted Control/Compute Unit combines functions that traditionally span multiple components, including the BMC, hardware Root of Trust, TPM, secure networking, firewall, and dedicated AI engines. The hardware can monitor system behavior and telemetry for threats such as ransomware, insider attacks, and side channel attacks while providing a trusted foundation for server management.

For AI infrastructure builders, that means security and management can begin inside the server itself, closer to the systems they are trying to protect.

Who they serve
Axiado is targeting cloud and AI data centers, server and system manufacturers, networking infrastructure, and 5G environments, with products designed for platforms including NVIDIA MGX and OCP based systems.

Explore Axiado to see how hardware anchored security is being built into the next generation of AI infrastructure.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

What's changing in global hiring?

Global hiring is changing fast. From AI's impact on HR to evolving compliance requirements and international expansion strategies, the rules are constantly shifting.

Oyster's events bring together HR leaders, founders, operators, and global employment experts to discuss what's working now—and what's coming next.

Whether you're hiring internationally today or planning for tomorrow, you'll walk away with practical insights you can actually use.

Explore Hiring Events

TAKEAWAYS FROM THE SILLICON VALLEY AI TOUR💰 Bay Area Startups Collectively Secured $2.1B+ in August Week 1

Sat, 08 Aug 2026 15:00:56 +0000

TAKEAWAYS FROM THE SILICON VALLEY AI STUDY TOUR

Last week, I delivered the opening keynote for the Silicon Valley AI Study Tour, a week-long immersion program organized by LinkSV and The Banking Academy, bringing 35 senior executives from 23 global banks into the heart of Silicon Valley to see firsthand what's actually happening in AI.
Their question was simple: what's real and what's hype? The answer wasn't what most of them expected.

Silicon Valley Isn't a Place. It's an Intersection.

Silicon Valley's real advantage is speed of information. This is where founders, operators, investors, and infrastructure builders physically overlap and share what they're learning in real time. That density of knowledge transfer doesn't exist anywhere else at the same pace.

The framing message: "The rate of learning must match the rate of change. Right now, this is the only place where that's consistently true."

For international financial leaders evaluating where to place AI bets, the takeaway was simple. You don't have to build here. But you need to be plugged in here.

My Keynote, August 03/2026

Three Layers Moving at Three Speeds

I broke the AI market into three layers, each on a different clock:

Software and applications - changing in days
Models and tooling - changing in months
Infrastructure - changing in years

Enterprise procurement, capital allocation, and regulatory frameworks all move on infrastructure timelines. But the technology they're evaluating shifts at software speed. For banks and institutional investors, the diligence cycle is often slower than the market itself.

Infrastructure Is the Bottleneck

The biggest constraints on AI scale right now are all at the infrastructure layer. GPU shortages. Energy shortages. Storage and memory limitations. And the sheer capital required to compete.

A viable neocloud needs roughly a billion dollars in GPUs and infrastructure. Right now we're advising on $100M+ raises that need to close within 90 days and billion-dollar targets over the next year just to secure enough data center capacity.

The training wave is already straining capacity. The inferencing wave, which will be bigger, hasn't fully arrived yet. Enterprise demand for private GPU clouds is slower than expected. But when it shows up, the gap gets wider.

Watch It Now

The Iron Man Suit

I also walked through how we run IgniteGTM internally. 14 people. 10 active AI agents. I call the operating model the "Iron Man suit."

The human operator wears the AI. We iterate workflows multiple times a day. Humans stay in control. The agents handle execution, research, drafting, coordination. But every output gets reviewed, revised, and redirected by a person.

We've replaced our conventional CRM with agentic workflows. Not by cutting people, but by making each person way more capable.

What the Bankers Took Home

The delegation left with a practical filter for evaluating AI opportunities:

Top-down projects driven by executive mandate or hype tend to stall
Bottom-up projects driven by real experimentation tend to stick

I also shared a note on hardware strategy: don't chase the newest GPU generation in emerging markets. Older but capable hardware is often more practical and deployable.

On trust especially around Chinese models and open source, enterprises will pay a premium for vendors they trust. Price and performance matter, but trust is the deciding factor in regulated industries.

AI is the most important discovery since fire. But the path to capturing that value runs through infrastructure, trust, and the willingness to learn faster than the market moves.

Check out our video library of over 300+panels, presentations and interviews

Get the MasterPass Today

Tickets Are Limited

Bay Area Startups Collectively Secured $2.1B+ in August Week 1

August started out slow, ending week one with just $2.1B in startup funding. The week's deals included fundings for AI infrastructure startups Buzz Solutions, Lumilens and Multibeam Systems.

SaaSpocalypse exits: There was one notable (but not for a good reason) acquisition this week – Bending Spoons' acquisition of Airtable for $1.28B. Their previous valuation was $11B in their 2021 Series F funding, $5.5B in the round before that and total funding of $1.36B. Not a win for the investors.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Aurelius Systems closed a $40M Series A, an autonomous, directed-energy technology company building the defense ecosystem of the future.
Sapiom closed a $35M Series A, provide the essential infrastructure (KYA, wallets, spending and usage controls, and multi-rail payments) to unlock secure, programmatic commerce.
Naive closed a $28.5M Series A, provides the full stack your agent needs to act in the real world — compute, payments, identity, incorporation, 10,000+ tools — in one config file.
Buzz Solutions closed a $20M Series A, an AI software company helping utilities modernize the electric grid through intelligent visual infrastructure management.
Avatar Robotics Co closed a $6.5M Seed, building the unlimited workforce for industrial labor.

Growth Stage:

Lumilens Inc. closed a $700M Series C, creates photonics products designed for high-performance computing.
Mariana Minerals closed a $310M Series B, a software-first, vertically integrated minerals company focused on the minerals critical to modern energy, AI, and defense technologies.
Multibeam Systems closed a $205M Series Unknown, helping accelerate semiconductor chip innovation with the first Multicolumn E-Beam Lithography system built for volume production.
Expedition Therapeutics closed a $115M Series B, a biotechnology company developing novel therapies for serious inflammatory and respiratory diseases.
Windborne Systems closed a $37M Series B, designs, manufactures, and operates the only constellation of smart, long-duration sensing balloons that power AI global weather forecasts.

Helios

Inference is becoming a much bigger infrastructure problem as AI moves into production. Every user request, agent action, generated image, and reasoning step consumes compute. At scale, teams have to think about throughput, latency, GPU availability, and what every inference request actually costs.

Companies are looking for a simple outcome: run models quickly and reliably without letting inference economics get out of control.

That is where Helios fits.

Who they are
Helios is an AI infrastructure provider offering GPU cloud, colocation, and serverless inference built around high density compute infrastructure.

CEO Jose Rojas & Chief Of Staff Michael Hawkins at AMD Advancing AI 2026

What they deliver
Helios gives teams access to open models including Kimi, DeepSeek, Whisper, and Flux through serverless inference with per millisecond billing. That inference platform sits alongside dedicated GPU infrastructure, including Blackwell generation systems designed for large scale reasoning and production inference.

Who they serve
AI companies, developers, and infrastructure teams that need production inference without building and operating the underlying GPU infrastructure themselves.

As token volumes grow and inference becomes a larger share of AI compute demand, the ability to deliver that compute efficiently will have a direct impact on what AI products can economically scale.

Explore Helios to see how they are approaching production AI inference.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Write docs 4x faster. Without hating every second.

Nobody became a developer to write documentation. But the docs still need to get written — PRDs, README updates, architecture decisions, onboarding guides.

Wispr Flow lets you talk through it instead. Speak naturally about what the code does, how it works, and why you built it that way. Flow formats everything into clean, professional text you can paste into Notion, Confluence, or GitHub.

Used by engineering teams at OpenAI, Vercel, and Clay. 89% of messages sent with zero edits. Works system-wide on Mac, Windows, and iPhone.

Try Wispr Flow free

WHY AI STRATEGY IS AN ORG DESIGN PROBLEM💰 Bay Area Startups Collectively Secured $14B+ in July MTD

Sat, 01 Aug 2026 15:00:54 +0000

Most of the conversation around AI and the workforce is trapped in two unhelpful extremes.

1) Fear driven thinking focuses on job loss, workforce reduction, replacement, and survival.

2) Shallow hype promises one human managing fifty agents with infinite scale, no friction, no organizational design, and no human complexity.

A more useful conversation starts with manpower design and how humans and agents should work together inside the same operating system.

Many companies still build teams using an old mental model centered on roles, titles, and headcount plans. We need a marketer. We need an operator. We need another coordinator. AI enabled environments require a broader view. Companies need functions and capabilities. They need judgment, orchestration, follow through, and creative problem solving to happen reliably inside the business. A human may provide that capability. An agent may handle part of it. A person equipped with agents may deliver it at a much higher level. Hiring is becoming one component of a larger system of execution.

There is a lot of noise from people claiming they have built giant agentic organizations. Dozens of agents are listed as proof of sophistication. Press even slightly and the model often falls apart. Where does judgment live? Who is orchestrating? What happens when the work changes shape? Agent count tells us very little about whether a company has built useful manpower.

Black box tools promise easy output. Push a button, get an answer, move on. Over time, that approach can separate people from the craft, workflow, and decision making logic that make a business run. AI strategies that turn employees into passive button pushers create fragile organizations.

This is where Conway's Law becomes essential, it tells us that organizations design systems that mirror their communication structures. The shape of a company influences how information moves, how decisions get made, and what kind of execution becomes possible. AI strategy therefore belongs inside the broader conversation about organizational design.

When decision making and orchestration flow through one or two people, additional agents can accelerate the existing bottleneck. The tools may be modern while the communication structure remains unchanged. In sports, formation determines which plays a team can run. Business operates the same way.

Before adding headcount, identify the function that is missing. Determine how much human judgment it requires and which parts an agent could handle. Find where decisions are bottlenecking. Understand how the new capability will change the communication structure of the company.

The answer may involve hiring, building an agent, giving an existing employee a stronger operating layer, or redesigning the structure around the work. Companies that figure this out early will become more efficient, adaptive, resilient, and human in the ways that matter most.

Check out our video library of over 300+panels, presentations and interviews

Bay Area Startups Collectively Secured $14B+ in July MTD

July started out slow, but accelerated to close on more than $14B in startup funding, with 25 megadeals (Safe Superintelligence with $5B, two more for $1B+, ) and a spike in seed fundings. More than 85% of July funding went into megadeals, with a fraction (less than 5%) going to seed rounds.

AI Infrastructure: As usage shifts to inferencing over training, we're seeing multiple startups launch to optimize for that – in silicon, software and infrastructure. The latest is Inferra AI, details below.

Exits: There were three notable transactions this week for more then $2.6B - Permiso Security for $199M, DroneDeploy for $845M and Anyscale for $1.65B.

For startups raising capital: Don't waste time on lists, find investors in companies like yours with LinkSV. Stay on top of who's raising, who's closing and who's investing with the Pulse of the Valley weekday newsletter. Founders get the newsletter, database and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

ChipAgents closed a $60M Series A, an agentic AI platform that dramatically shortens design and verification cycles in semiconductor workflows.
Revoy closed a $27M Series A, electrifying road freight at lower costs than diesel.
Inferra AI closed a $19.3M Seed, building infrastructure specifically designed for high-performance inference.
Aranya closed a $11.3M Series A, converts bare metal GPU infrastructure into highly custom production clusters in under 48 hours
Pilot Protocol closed a $4.5M Seed, building a peer-to-peer networking protocol where AI agents discover, trust, and transact with one another.

Growth Stage:

Safe Superintelligence closed a $5B Corporate, has one goal and one product: a safe superintelligence.
Antora Energy closed a $550M Series C, thermal batteries convert low-cost, intermittent renewable electricity into reliable industrial energy.
Eliyan closed a $145M Series C, their advanced connectivity technologies address critical bandwidth, power efficiency, and scalability challenges across modern compute infrastructure.
PSIquantum received a $125M Grant, building the world’s first useful quantum computer.
Regatta closed a $68M Series B, a transactional and analytical database that executes parallel queries and helps deliver apps with less code in a faster time.

Inflect

Buying digital infrastructure is still surprisingly manual.

A company can compare flights, hotels, and rental cars in minutes. Pricing a network connection or finding the right colocation provider often means emails, spreadsheets, phone calls, and waiting on quotes.

That process hasn't kept pace with the industry it supports.

Left To right: VP of Partnerships Mike Hiatt, CEO Mike Nyugen, Bill Barry, VP of Sales Josh Pittman, Director of Strat Solutions Cliff Martin

Inflect is building a marketplace for digital infrastructure, giving buyers a single place to research, compare, and source connectivity, cloud, and colocation services from providers around the world. The platform brings together pricing, availability, and infrastructure data that has traditionally been scattered across dozens of vendors and conversations.

As demand for AI infrastructure continues to grow, speed matters. The companies that can evaluate options and deploy infrastructure faster gain an advantage long before hardware is installed.

Follow the team at Inflect and see how they're simplifying infrastructure procurement.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

The AI Agent You Can Trust

The best assistants don't multitask their attention across a hundred tools. Neither does Catch. It's an AI agent that focuses on one thing — the admin work you'd rather not touch — and does it exceptionally well.

Scheduling, flights, restaurants, follow-ups, vendors, clients. You hand it over; Catch handles the back-and-forth and comes back with it done.

No context-switching. No dropped balls. Just your admin, quietly cleared — so your focus stays on the work only you can do.

Meet the agent built for admin, and it'll be ready to work before your next meeting.

Get started at catchagent.ai — and give your attention back to what matters.

Start now for free

KEY TAKEAWAYS FROM AMD ADVANCING AI DAY💰 Bay Area Startups Collectively Secured $7.6B+ in July MTD

Sat, 25 Jul 2026 16:00:00 +0000

AMD just laid out its roadmap for the next era of AI compute and if you build, operate, or buy AI infrastructure, here's what matters.

1. The data center is the new unit of design

Dr. Lisa Su made it clear: the days of optimizing individual accelerators in isolation are over. Power, cooling, memory, and networking are now first-class design constraints. If your rack architecture doesn't account for all of them together, you're leaving performance on the table.

AMD CEO Dr Lisa Su at Advancing AI 2026

2. Helios is AMD's answer to rack-scale AI

AMD introduced Helios, a modular platform that integrates Instinct accelerators, EPYC CPUs, and Pensando networking at the rack level. The pitch: better performance per watt and simpler deployment across large clusters. For infrastructure operators, this is AMD signaling it wants to compete on full-stack system design, not just chips.

3. The accelerator market is heading past $200B

They project the data center AI accelerator market will grow at 50%+ CAGR. Demand is coming from more than hyperscalers now, enterprises, sovereign AI programs, research institutions, and emerging neoclouds are all driving spend. The buyer base is fragmenting, which means the infrastructure that serves them has to be more flexible.

4. The hardware roadmap extends through 2028

MI500 in 2027. MI600 in 2028. Both built on HBM4 memory architectures for larger training and reasoning workloads. For anyone planning data center capacity, power budgets, or capital deployment this is the planning horizon AMD is giving you.

5. Software openness is the competitive wedge

AMD is leaning hard into an open ecosystem play. In a market where CUDA lock-in is the default, AMD is betting that flexibility and portability will matter more as the buyer base diversifies beyond the hyperscaler top 5.

AI infrastructure is now a systems game; power, cooling, memory, networking, and software all have to move together. they’re positioning itself as the platform for operators who need choice, efficiency, and a roadmap they can plan around. Whether you're a hyperscaler, a neocloud, or an enterprise building out AI capacity, this is the competitive landscape shifting in real time.

Check out our video library of over 300+panels, presentations and interviews

We’re proud to be a community partner of the AI Summit at Stanford a three-day residential gathering for AI founders, investors, technologists, and researchers.

The summit runs July 30 to August 1 on Stanford’s campus.

Members of the IgniteGTM community receive 50% off tickets with code IGNITEGTM. Space is limited.

Use code IGNITEGTM for 50% off

Bay Area Startups Collectively Secured $7.6B+ in July MTD

Funding activity stayed steady this week, but the median size of the deals went down from $24M to $11.5M. There were only five megadeals and the round size dropped from there, driven by a flurry of seed rounds that made up almost 50% of this week's fundings.

Exits: Out of ten acquisitions this week, only one was notable - Tempus acquiring Personalis for $1.5B. Scribe Therapeutics made their public debut in the first gene-editing IPO in more than two years, raising $129M after upsizing shares to 8.5 million and pricing at $15, the top of their range. The stock (SCTX) opened on their first day of trading at $25 and closed at $21.65.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, new tech and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Natural closed a $30M Series A, building the financial rails for an agentic future where autonomous agents initiate, authorize, and execute payments.
Elio closed a $21M Series A, their sensors let AI direct what to capture in real time.
SkyPilot closed a $20M Seed, AI Compute Platform for frontier AI teams.
Runta closed a $20M Seed, the execution layer for AI agents.
Infinity.inc closed a $15M Seed, building the software layer that makes any chip competitive for AI inference.

Growth Stage:

Meshy AI closed a $400M Series B, an artificial intelligence platform that enables users to generate and edit 3D models from text descriptions, images, and conversational prompts.
Sila Nanotechnologies closed a $300M PE round, a next generation battery materials company dedicated to accelerating energy transformation.
Etched closed a $300M Series C, builds frontier inference clusters designed to make AI inference dramatically faster, cheaper, and more abundant.
Candid Health closed a $120M Series D, on a mission to rebuild the financial infrastructure of American healthcare
Glow Security closed a $100M Series B, the Endpoint AI company with an AI-powered operating model built on prevention, not reaction.

AI inference is becoming more specialized. As models grow larger and AI agents generate more tokens, infrastructure teams are looking beyond one-size-fits-all architectures. The goal isn't simply adding more accelerators. It's matching the right hardware to the right workload.

That's the idea behind the new partnership between AMD and Cerebras.

AMD and Cerebras have announced a joint inference solution that combines AMD's Helios rack-scale AI infrastructure with the Cerebras Wafer-Scale Engine. The architecture pairs high-throughput processing with ultra-low-latency inference, allowing different parts of the workload to run on the hardware best suited for the job.

Rather than relying on a single platform for every stage of inference, operators are increasingly designing specialized, composable systems that optimize for performance, efficiency, and scale.

Learn more about how AMD and Cerebras are approaching the next generation of AI inference infrastructure.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Same AI software. Wildly different results.

Every company in this dataset bought the same AI capabilities. The difference in results came down to one thing: whether someone inside CX owned it.

One beauty retailer made 202 workflow updates in 30 days — refining as policies changed and new questions came in. Companies without a named owner saw performance stall or decline.

Read the data on what separates AI deployments that work from the ones that stall, and the four questions worth asking before your next AI investment.

See the data

🗂️Infrastructure Masons Report Out Now💰 Bay Area Startups Collectively Secured $6.1B+ in July MTD

Sat, 18 Jul 2026 15:29:22 +0000

The 2026 iMasons State of the Digital Infrastructure Industry Report is now live. Their document captures an industry expanding at a pace rarely seen outside major waves of industrial development. Drawing on input from nearly 2,000 industry leaders, the report places AI infrastructure inside a much larger system shaped by power, policy, labor, capital, community support, and long-term demand for digital services.

The scale of the buildout is difficult to overstate. Digital infrastructure has become a global industry spanning thousands of data centers, with most planned capacity still waiting to be built.

AI demand is driving larger projects on shorter timelines. Development schedules that once stretched 24 to 36 months are being compressed to 12 to 18 months, while individual AI campuses are growing from hundreds of megawatts to multiple gigawatts. That combination of speed and scale intensifies pressure across the entire supply chain. Equipment backlogs are growing, construction timelines are slipping, and local labor markets are being asked to support projects that may require more than 10,000 workers on-site each day for several years.

Power remains the industry's defining constraint. Delayed grid connections are pushing developers toward bridge power solutions while utilities race to expand generation and transmission capacity.

Policy now touches almost every part of the buildout. It shapes data protection, power access, workforce development, permitting, environmental standards, water use, and the way communities experience new development. Regions with stable regulation and clear approval pathways are gaining an advantage, particularly as traditional data center hubs face tighter power markets and longer development cycles.

Community support has become a material factor in project viability. More than $150 billion in U.S. data center projects were delayed or blocked in 2025, reinforcing how local concerns increasingly shape where and how infrastructure gets built.

Geography is becoming a competitive advantage. Regions able to align power, regulation, capital, connectivity, and community support are attracting the next wave of AI investment.

The report leaves a clear impression of where the industry is heading. AI infrastructure is becoming a coordinated industrial system. Data centers sit at the center, but the outcome depends on transmission lines, tradespeople, utilities, regulators, investors, and communities moving together. The regions that align those pieces will shape the next phase of global AI development.

Explore the complete 2026 State of the Digital Infrastructure Industry report for the full research, data, and regional analysis.

Download the report

Check out our video library of over 300+panels, presentations and interviews

The infrastructure powering that shift takes center stage at AMD Advancing AI.

Join us July 22-23 at Moscone West in San Francisco.
Register today while space is still available.

Apply Today

Bay Area Startups Collectively Secured $2.7B+ in July MTD

July just got busy, with Fireworks AI's $1.5B Series D the largest of nine megadeals, spread across AI, applied AI, fintech and biotech.

Not NVIDIA: Coreweave pioneered the use of GPUs as collateral in corporate financing. AI inference startup General Compute is now the first to use the SambaNova inference specific RDUs as collateral for a $400M loan. The loan is on top of the $15M seed round CEO Finn Puklowski raised in May to build an inference neocloud with the SambaNova silicon. The RDUs are power-efficient and don't require water cooling, so they can be deployed more quickly across a larger variety of data centers. General Compute says their neocloud will provide inferencing 16 times faster than a GPU cloud.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, new tech and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

TYLsemi closed a $43M Series A, redefining how AI infrastructure silicon is built, delivering standards-based, production-ready chiplets purpose-built for modern multi-die systems.
Rime Labs closed a $24M Series A, trusted enterprise voice AI platform, built on a linguistics-first foundation to deliver the world’s leading conversational speech models.
Flow Labs closed a $14M Seed, AI-powered platform, generating the most accurate transportation data available today, optimizing traffic flow across an entire city in seconds.
Aina closed a $5.5M Seed, consumer hardware company building a general-purpose AI interface beyond phone touchscreens and keyboards.

Growth Stage:

Fireworks AI closed a $1.5B Series D, platform for specialized intelligence, enabling companies to train and serve models tailored to their own data, workflows, and use cases.
Chai Discovery closed a $400M Series C, building a world where new medicines are designed with the precision, speed and scale of modern engineering.
Spectro Cloud closed a $100M Series D, helps platform teams and cloud providers modernize and manage infrastructure for the AI era without adding more tools or operational complexity.
Quadric.io closed a $18M Series C, the inference engine inside on-device AI chips.
Our Friend Dean Nelson

Dean Nelson Moderating at AIS5

The AI industry spends a lot of time talking about GPUs.

Dean Nelson has spent his career thinking about everything around them.

Long before AI infrastructure became front-page news, Dean was working on the problems that determine whether large-scale computing is actually possible, power, cooling, reliability, and the people needed to build and operate it.

That perspective led him to found Infrastructure Masons (iMasons), an organization built on a simple idea: some of the industry's biggest challenges are too large for any one company to solve alone.

Today, iMasons brings together leaders from across the digital infrastructure ecosystem to tackle issues ranging from workforce development and sustainability to power and industry standards. At a time when AI is putting unprecedented pressure on physical infrastructure, those conversations have become increasingly important.

If you care about where the infrastructure industry is headed, Dean is one of the people worth following.

Follow Dean Nelson on LinkedIn:
https://www.linkedin.com/in/deannelson/

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Thinking about hiring globally? Start with an EOR.

The best person for your next role might not live near your office—or even in the same country.

More companies are realizing they don't need to open entities everywhere just to access global talent. Instead, they're using EOR to hire internationally faster, stay compliant, and avoid building local infrastructure before they're ready.

Oyster's EOR helps companies hire, pay, and support employees in 180+ countries while Oyster handles payroll, compliance, taxes, and local employment requirements.

See how it works

🧮Real Budgeting For AI 💰 Bay Area Startups Collectively Secured $2.7B+ in July MTD

Sat, 11 Jul 2026 15:38:54 +0000

Spend enough time reading AI headlines and you'd think enterprises are slamming on the brakes.

Stories about developers burning through token budgets, companies imposing monthly spending caps, and employees losing access to premium models have created the impression that enterprise AI spending is running into a wall.

SemiAnalysis recently spoke with more than 50 enterprise customers to understand how organizations are managing AI token spend. What emerged wasn't a picture of companies spending less on AI. It was a picture of companies becoming much more deliberate about where AI creates value. Budgets are becoming part of normal operations, even as AI investment continues to grow.

What's interesting is that there doesn't appear to be a playbook yet. Some organizations cap employees at a few hundred dollars a month. Others allow several thousand depending on role. Some have no formal limits at all. The number itself seems less important than the exercise. Enterprises are trying to understand how much AI different jobs actually need and where additional spending continues to improve business outcomes.

There's another reason those conversations are becoming more important.

The instinct is to assume that as AI models become faster and less expensive, infrastructure costs should become easier to manage. Instead, the opposite is happening. Lower token prices are making AI economical for more teams, more applications, longer context windows, and increasingly autonomous agents. Every improvement in efficiency creates another reason to use AI more often.

Jevon’s Paradox In Action

Economists have a name for this phenomenon. Jevons paradox describes what happens when efficiency improvements increase total consumption instead of reducing it. WEKA's Val Bercovici has argued that AI is becoming one of the clearest modern examples. Cheaper tokens don't reduce infrastructure demand. They unlock new workloads that were previously too expensive to justify.

That dynamic helps explain why enterprises are spending so much time thinking about governance. The challenge isn't simply controlling token spend. It's deciding which workloads require frontier reasoning models, which can run on smaller or open-weight alternatives, and how to keep increasingly autonomous systems from consuming expensive compute where it adds little value.

The organizations making the largest AI investments tend to have one thing in common. They're spending where AI directly affects output. Engineering teams, data scientists, and technical researchers consistently receive larger budgets because their work benefits from iterative reasoning, code generation, and large-context analysis. Other functions often don't require the same level of compute. Instead of treating AI like another software license, companies are beginning to allocate intelligence much the way they allocate any other strategic resource.

The same thinking is showing up in infrastructure decisions. Many organizations are changing their default models, reserving premium reasoning models for complex work while routing routine tasks to lower-cost alternatives. Others are investing in better caching and memory architectures so they aren't repeatedly paying to generate the same outputs. Those decisions reduce unnecessary spending without reducing AI adoption.

The productivity gains explain why budgets continue to expand despite greater scrutiny. SemiAnalysis highlighted examples ranging from recruiting workflows that were cut dramatically to analytical work that dropped from days to hours. Those kinds of improvements change the economics. Once a workflow demonstrates measurable value, the question shifts from "Should we spend money on AI?" to "Where should we deploy more of it?"

Enterprise AI hasn't reached a spending ceiling. It's reaching a planning phase. Budgets, model routing, memory architecture, and infrastructure strategy are all becoming part of the same conversation because they're responding to the same reality. As AI becomes cheaper, organizations don't consume less intelligence. They find more places to apply it. The companies that recognize that dynamic early will be better positioned to build infrastructure for where enterprise AI is going, not where it has been.

Check out our video library of over 300+panels, presentations and interviews

The infrastructure powering that shift takes center stage at AMD Advancing AI.

Join us July 22-23 at Moscone West in San Francisco.
Register today while space is still available.

Apply Today

Join In

Bay Area Startups Collectively Secured $2.7B+ in July MTD

July continued its slow start, with just two megadeals, SambaNova's $1B Series F the largest. But the number of deals was half the usual and begs the question, is this the summer slowdown, or the lull before the storm?

AI Revenue Gap: Sequoia's David Cahn was one of the first people to dig into projected AI infrastructure spend and the revenue needed to pay for it. Three years ago - when NVIDIA was reporting $50B in annual GPU revenue - he pegged it at $200B in AI revenue. Today, he's estimating that AI infrastructure spend in 2026 will hit $1.5T - and puts the corresponding AI revenue number at $300 trillion. Although OpenAI and Anthropic are both reporting ARR in the tens of billions, that leaves a big gap in revenues as the hyperscalers continue to build and take on more debt.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, new tech and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Prime Intellect closed a $130M Series A, makes frontier AI training accessible to every company.
Bespoke Labs closed a $40M Series A, a research lab building the environments and infrastructure that make AI agents reliable.
Hostie closed a $12M Series A, the AI-powered Virtual Concierge for restaurants.
Piq Energy closed a $5M Seed, helps companies connect projects to the grid faster with an agentic grid planning platform.

Growth Stage:

SambaNova Systems closed a $1000M Series F, developer of the SN50 RDU (Reconfigurable Dataflow Unit), an AI processor purpose-built for agentic AI inference at lower cost, and higher energy efficiency than GPUs.
Ollama closed a $65M Series B, makes open models easy, efficient, and safe to use on a user's own device or in Ollama’s cloud.
Super closed a $65M Series D, the all-in-one app to save, earn and put more money in your pocket.
Hippo Harvest closed a $30M Series C, a leafy greens company growing USDA-certified organic produce in robotics- and machine learning-powered greenhouses.

NeoCloud Council Executive Dinner

Some of the most valuable conversations in AI infrastructure never happen on stage.

They happen around a dinner table.

The NeoCloud Council Executive Dinner brings together founders, cloud providers, enterprise leaders, investors, and infrastructure operators for an evening designed around discussion instead of presentations.

Hosted during AMD Advancing AI Week, the dinner is intentionally small. It's creating the kind of room where people can compare notes on what's actually happening across AI infrastructure, from deployment challenges and customer demand to where the market is heading next.

The people building this industry rarely get the chance to talk candidly with peers facing the same decisions. That's what this dinner is for.

If you're helping shape the next generation of AI infrastructure, this is a conversation worth being part of.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Speak naturally. Send without fixing.

Wispr Flow turns your voice into clean, professional text you can send the moment you stop talking. Not rough transcription you have to clean up. Actual polished text — ready for email, Slack, or any app.

Speak the way you think. Go on tangents. Change your mind mid-sentence. Flow strips the filler, fixes the grammar, and gives you text that reads like you spent five minutes writing it.

89% of messages sent with zero edits. Millions of professionals use Flow daily, including teams at OpenAI, Vercel, and Clay. Works on Mac, Windows, and iPhone.

Start flowing free

🦸 AI Agents: Iron Man suits & Superteams 💰 Bay Area Startups Collectively Secured $22B+ in June and $330B YTD

Sat, 04 Jul 2026 15:00:00 +0000

How AI agents are helping small teams operate with the reach and consistency of much larger organizations.

AI conversations often focus on models, tools, and automation.

What interests us more is what happens to the people using them.

Over the last several years, we've been building and refining an internal system of AI agents across IgniteGTM. Along the way, we found ourselves coming back to the same analogy.

The Iron Man suit.

Tony Stark built the suit to extend his reach. It gave him access to capabilities he would not otherwise have. Strength, speed, protection, live analysis, and situational awareness all became available while he remained firmly in control.

That idea became the foundation for how we think about AI.

The Iron Man Principle

We're a lean agency that produces large-scale AI infrastructure events, runs GTM advisory engagements, and manages sponsorship pipelines across dozens of enterprise partners. The kind of work that often requires significantly larger teams.

As we looked at how AI could support the business, we became less interested in automating individual tasks and more interested in expanding what people were capable of accomplishing.

Every role eventually runs into limits. There are only so many customers you can actively support, opportunities you can pursue, projects you can oversee, and commitments you can keep moving forward at the same time.

The agents help push those limits outward.

They assist with preparation, research, coordination, follow-up, documentation, and information sharing across the organization. Conversations remain connected to projects. Decisions remain connected to action items. Context remains available long after a meeting ends.

The impact is cumulative.

A salesperson can effectively manage more relationships without sacrificing responsiveness. A project lead can maintain visibility across more moving parts without losing track of details. A leadership team can operate across a broader range of initiatives without becoming buried in coordination.

The relationships still belong to people. The decisions still belong to people. The strategy still belongs to people.

What changes is the amount of responsibility, complexity, and opportunity each person can successfully handle.

Last Suit You'll Ever Wear

The agents operate in the same Slack channels, inboxes, and project boards as the rest of the team. Nothing discussed in a call gets lost.

Agents monitor conversations, summarize information, surface action items, and maintain continuity across projects. Decisions, contacts, notes, and project history remain connected as work moves through the organization.

The key design principle is dual-write everything. Decisions, contacts, and notes are saved both locally and to a shared database. That reduces the risk of lost context and single points of failure.

The system continues to evolve as the business evolves. New workflows are introduced when they solve a real problem. Existing workflows are refined as requirements change. Some are removed entirely.

The objective is not to build the largest collection of agents possible. The objective is to create systems that make people more effective, more capable, and better equipped to execute and move at the speed of the spoken word.

The Human-in-the-Loop Advantage

AI agents work best when paired with human oversight.

Every draft gets reviewed by a person before it sends. Every outreach sequence gets approved. Every meeting brief is a recommendation, not a decision.

The agents gather information, maintain continuity, and support execution. People remain responsible for judgment, prioritization, relationships, and strategy.

That distinction matters.

The value is not found in handing responsibility to software. The value comes from enabling people to operate with a wider field of view and a greater capacity to act.

As responsibilities grow and organizations become more complex, those advantages compound. People can remain engaged across more initiatives, support more customers, manage more projects, and maintain higher levels of consistency than would otherwise be practical.

Strategic Imperative: Why Start Now

Most organizations already have access to the tools, but the bigger challenge is learning how to incorporate them into everyday work.

For some teams, the starting point may be meeting preparation. For others, it may be documentation, research, customer follow-up, project management, or internal communication.

The organizations making the most progress are learning through implementation. They are refining workflows, improving information flow, and discovering where AI can expand the capabilities of the people already on the team.

The Suit Is the Advantage

The next several years will not be defined by access to AI. Most organizations will have access to similar technology. The difference will come from how effectively they use it.

Organizations do not need a massive AI deployment on day one. They need a workflow worth improving, a team willing to experiment, and a commitment to continuous refinement.

The companies that succeed will be the ones that find ways to expand the capabilities of their people.

That's how we've approached it at Ignite.
And that's why we call it our Iron Man suit.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 3rd, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today

// UPCOMING EVENTS

Apply Today

Join In

Bay Area Startups Collectively Secured $22B+ in June and $330B YTD

The first half of 2026 brought in $330B in funding to SV startups, almost 3X the 2021 full year funding total, when interest rates were close to zero and startup investments flooded the valley. July started off quietly, with less than $2B in fundings, and 95% going to AI infrastructure.

Joining the club: Anthropic is planning to develop a custom AI server chip, joining OpenAI (Jalapeno), Amazon (Trainium), Microsoft (Maia) and Google(TPU). There's a common interest among the hyperscalers in custom silicon that can lessen dependence on NVIDIA and increasing inference efficiency while reducing power requirements. Groq was in that category before NVIDIA did a reverse acquihire to get their IP; they've now pivoted to providing an inference cloud platform.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Tripo AI closed a $150M Series A, a global artificial intelligence company building AI 3D foundation models and world models.
CarbonSix closed a $40M Series A, a Physical AI company developing deployment-ready robotic intelligence and automation solutions for manufacturing.
Oxmiq Labs closed a $35M Series A, develops licensable compute IP and adaptive AI infrastructure software for developing and deploying custom AI compute.
Omen AI closed a $31M Series A, Omen uses continuous fluid analysis to read the health of data center equipment in real time.
Pocket Inc. closed a $11M Seed, an AI hardware company building a dedicated device for memory, recall, and post-conversation workflows.

Growth Stage:

Together AI closed a $800M Series C, the AI Native Cloud, combining open-source models, high-performance infrastructure, and frontier research in AI efficiency and scalability.
Twelve Labs closed a $100M Series B, the world's most powerful video intelligence platform, that enables machines to perceive, understand, and reason about video the way humans do.
Materna Medical closed a $5M Series B, a novel OBGYN platform company with core technologies addressing unmet needs in women's pelvic health.

Meet Chiri

Every company has a few people who keep everything moving. They know which customer requests need special handling, which process breaks under pressure, and which spreadsheet everyone depends on even though it isn't documented anywhere.

That knowledge is incredibly valuable, and incredibly fragile.

When those people are overloaded, on vacation, or leave the company, the business feels it immediately.

Matthew Wimberly, Co-founder & CTO of Chiri AI at Enterprise AI & HPC Summit NYC

Chiri is building around that reality. Their platform helps organizations capture how work actually gets done and put that knowledge to work inside the systems teams already use. The goal is straightforward: make critical knowledge easier to access, easier to share, and less dependent on a handful of individuals.

As more organizations push AI into day-to-day operations, the companies that can operationalize their expertise will have a significant advantage.

Because scaling a business often comes down to one question:

Can your best people make everyone else better?

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Six people doing the work. Your headcount is one.

Your finance close runs in #finance. Stripe and QuickBooks reconciled, runway updated, posted Sunday night without you asking.

Engineering review lands in #eng. Viktor pulled the open PRs, left comments on auth-refactor, flagged a dependency blocking api-pagination.

Campaign brief lands in #growth: Meta CPA up 18%, recommendation to pause broad match, a draft landing page already deployed for the variant test.

You hired him on day zero. He lives in Slack and Microsoft Teams alongside your contractors and investors, connects to 3,000+ tools, pushes back when you ship something dumb.

"Viktor is now an integral team member, and after weeks of use we still feel we haven't uncovered the full potential." Patrick, Director, Yarra Web.

Get Started for Free

🗄️Takeaways From Supermicro's Open Storage Summit 💰 Bay Area Startups Collectively Secured $21B+ in June MTD

Sun, 28 Jun 2026 15:01:06 +0000

Left to right: Rob Strechay (theCUBE), Bill Miller (MinIO), Kurt Kuchein (Hammerspace), and William Li (Supermicro

Enterprise AI's Hardest Problem Lives in the Data Layer

At Supermicro's Open Storage Summit Reception on June 24 in Mountain View, we helped bring together leaders from across the infrastructure stack to tackle one question head-on: What is the single biggest factor preventing organizations from scaling AI into core business processes?

Rob Strechay (theCUBE) moderated, alongside William Li (Supermicro), Kurt Kuchein (Hammerspace), and Bill Miller (MinIO), digging into the operational realities that still separate AI pilots from production deployments. The conversation reinforced something we've been tracking closely across the community; the scaling challenge has moved well past compute and models. The hardest problems now live in the data layer.

The POC Trap

Bill Miller framed the core tension clearly. In a proof of concept, the data is curated, the environment is controlled, and the use case is narrow. In production, none of that holds. Enterprise data is fragmented across clouds and on-prem systems, riddled with stale records, and governed by access policies that were never designed for AI workloads. A demo that dazzles a boardroom can collapse the moment it meets real-world complexity.

This is a pattern we see playing out across the infrastructure landscape and the panel treated it as a structural problem demanding architectural solutions.

Three Walls Between Pilot and Production

Miller broke the scaling challenge into three concrete barriers. First: access to all relevant data, particularly the sensitive on-prem "crown jewel" datasets that organizations are most reluctant to expose to AI pipelines. Second: the explosive growth of KV cache requirements at production scale. Third: the massive gap between internal deployments serving hundreds of employees and external deployments handling millions of transactions.

These are the operational categories infrastructure architects and enterprise leaders need to plan around now and the kind of specific, execution-level detail that drove this conversation.

Another Amazing Partner Event

Token Value Over Token Volume

One of the strongest threads to emerge was the distinction between how much data a model consumes and whether that data is actually useful. The panel argued that enterprises should focus less on throughput and more on whether models are ingesting relevant, current, authorized information. When an LLM treats an archived project plan from 2019 with the same weight as a live production dataset, output quality degrades and user trust follows.

That trust problem is where technical debt meets change management. Hallucinations and unreliable outputs stall adoption across the organization. Once confidence breaks, getting business units to rely on AI-driven workflows becomes an uphill fight.

The Unglamorous Infrastructure Layer

Kuchein and Li both pointed to a reality that rarely makes keynote slides: while the compute side of the AI stack has become increasingly streamlined, the storage and data management layer remains fragmented. Enterprises are still stitching together compliance tools, classification engines, storage orchestrators, and governance frameworks. The next wave of enterprise AI value may come from making that data layer manageable and coherent.

Open protocols figured prominently. Iceberg as a unified catalog layer, NFS for interoperability, and open-weight models were all positioned as mechanisms to reduce lock-in and bridge the structured-unstructured divide. Lakehouses combining the strengths of data warehouses and data lakes emerged as the architectural pattern for AI-ready enterprises.

Where the Conversation Goes From Here

This panel reinforced a thesis we keep coming back to in every conversation with infrastructure operators: enterprise AI adoption stalls when the data layer is incomplete, untrusted, or too complex to operationalize. Fix the data foundation governance, access, quality, hybrid reach and the rest of the stack becomes dramatically more usable.

The organizations that figure this out first will deploy AI that their people actually trust enough to use. That's where this conversation is headed, and we'll keep driving it.

Join the conversation

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today

// UPCOMING EVENTS

Bay Area Startups Collectively Secured $21B+ in June MTD

Silicon Valley startup funding activity spiked this week to $5.4B, with twelve megadeals – two of them each one billion or more – BaseTen with a $1B Series F and AppsFlyer with a $1B Series E.

Exits, M&A: M&A was the big story this week, with three billion-dollar plus acquisitions. $7B for Synaptics, $10.9B for Apogee Therapeutics and $4B for Modular, plus several smaller ones came to more than $22B in total, the most for any week in 2026.

For startups raising capital: Don't use lists, find investors in companies like yours with LinkSV. Stay on top of who's raising, who's closing and who's investing with the Pulse of the Valley weekday newsletter. Founders get the newsletter, database and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Mirendil closed a $200M Seed, an AI research lab building self-accelerating systems that turn compute into scientific and engineering breakthroughs.
Upscale AI, Inc closed a $190M Series A, a high-performance AI networking company accelerating AI democratization through open-standard, full-stack, turnkey solutions.
Sail Research closed a $80M Series A, the infrastructure company purpose-built for long-horizon AI agents.
Coval closed a $28M Series A, accelerates AI agent development with automated testing for chat, voice, and other objective-oriented systems.
Bolt Graphics closed an undisclosed amount Seed, a semiconductor startup building the fastest and most efficient graphics processors.

Growth Stage:

BaseTen closed a $1.5B Series F, builds systems software that runs the entire workload for AI applications—from GPUs and auto-scaling to observability, billing & developer tools.
Groq closed a $650M Series F, operates the inference cloud infrastructure that runs real-time AI.
AirWallex closed a $320M Series H, global financial platform, building the future of global banking for a borderless, real-time, intelligent economy.
Assort Health closed a $120M Series C, we manage the entire patient journey for providers, from call center automation and form intake to referrals, document processing, outreach, and payments.
Patronus AI closed a $50M Series B, a frontier lab developing simulation research and infrastructure to accelerate progress toward human-aligned AGI. We are training the first world models to simulate digital workflows.

When enterprises move AI from pilot to production, the conversation shifts fast from models and frameworks to hardware, logistics, and capital. That's where Avnet operates.

Founded in 1921 and now a $26.5 billion global technology distributor (Nasdaq: AVT), Avnet has spent over a century connecting enterprises with the components and infrastructure they need to build at scale. They were Intel's first distributor in 1973. Today, that same distribution muscle is pointed squarely at the AI infrastructure buildout.

For organizations deploying GPU clusters, high-density compute, and AI-optimized storage, Avnet solves the problems that sit between "we have a vendor" and "we have a running data center." Inventory management across global supply chains. Financing options that make large-scale hardware procurement possible without crushing capital budgets. Technical integration support that bridges the gap between component selection and production-ready deployment.

Paul Oldham & David Ellis from Avnet presenting at OSS

As a co-sponsor of Supermicro's Open Storage Summit Reception where Avnet delivered the keynote ahead of the enterprise AI panel their presence underscored a critical reality: the AI infrastructure era needs more than chip makers and software vendors. It needs distributors who can move hardware at scale, manage complex procurement cycles, and help enterprises operationalize their AI ambitions.

Avnet is that bridge and they've been building it for over 100 years.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Your membership list is full. Your calendar isn't.

You have hundreds of members needing tune-ups twice a year.

Scheduling them means multiple calls and weeks of manual outreach that never fully gets done.

The ones who don't get reached fall through the cracks. Those are members and recurring revenue quietly walking out the door.

Podium's AI Membership Coordinator texts every member, books the appointment, follows up automatically, and handles renewals.

It works your entire list proactively so your slow season fills before it starts.

No manual outreach. No dropped members. No empty shoulder season.

Make the most of your membership list.

Fill My Seasonal Calendar

5️⃣ Enterprise AI Trends That Defined HPE Discover 2026💰 Bay Area Startups Collectively Secured $16B+ in June MTD

Sat, 20 Jun 2026 18:39:17 +0000

HPE Discover, Las Vegas 2026

Enterprise AI is moving from experimentation to industrialization.

That was the clearest signal coming out of HPE Discover this week. Not any single product announcement. What stood out was where enterprises are putting budget, talent, and executive attention over the next several years.

Five themes kept showing up.

1. AI Is Becoming an Infrastructure Problem

For the last two years, most AI conversations revolved around models, copilots, and chat interfaces. That conversation is maturing. Attention is shifting toward data pipelines, networking, storage, governance, security, and inference infrastructure.

The Path to the Agentic Enterprise - Fidelma Russo Keynote

That changes who wins. Having the best model matters. But deploying AI reliably across thousands of users, systems, and workflows may matter more.

2. Networking Is Back in the Spotlight

Rami Rahim's networking keynote revolved around a simple idea: “Architecting AI starts with your network.”

Two years ago, networking barely entered the AI discussion. As a long time networking guy, I was perplexed to see it pushed aside as ‘dumb plumbing’. Today, distributed inference, AI factories, and massive data movement are pushing it back to the center. The cloud era revolved around compute. Big data elevated storage. AI is making networking strategic again. Moving data efficiently may become just as important as generating intelligence.

Watch Rami Rahim Keynote

3. The Enterprise Is Becoming "Agentic"

Russo described what she called the "distributed agency enterprise." The phrase matters less than the operating model behind it. AI is spreading across workflows, infrastructure, and business functions instead of living inside a single application. As Russo explained, systems are evolving to: "Observe, reason and act." That's a meaningful shift from where enterprises were just two years ago.

The question used to be: "How can AI help employees?"

Increasingly, the question is: "Which workflows should AI own?"

4. Hybrid Won

The assumption that everything eventually ends up in public cloud feels increasingly outdated. Enterprise AI will span on-prem environments, colocation, sovereign infrastructure, and public cloud.

Not because organizations are resisting change. Because latency, security, regulation, and economics demand it. Hybrid isn't the compromise and for many enterprises, it's the architecture.

5. Operationalizing AI Is the Real Challenge

Most buyers have moved beyond debating whether AI matters. The focus now is operating it at scale. Russo described the goal as "Closed-loop operations" that drive continuous observation, reasoning, action, and validation. She also offered perhaps the best measurement framework of the week:

Three Signals Worth Watching

1. Infrastructure spending is accelerating

Budgets are moving beyond pilots and into production. The money is following deployment.

2. AI factories are becoming a category

Treating AI as an industrial system, complete with repeatable processes, governance, and operations, is no longer theoretical. Multiple ecosystem players are converging around the same idea.

3. Token economics are heading to the boardroom

AMD CIO Hasmak Ranjan shared a useful example. At roughly $200 in tokens per employee per week, AI spending approaches $10,000 per employee annually. Those numbers get executive attention quickly.

The Takeaway

The biggest signal from HPE Discover was a mindset shift.
The conversation has moved beyond experimentation.
Enterprises are now figuring out how to run AI securely, economically, and at scale.

And that means the infrastructure underneath has to keep up.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today

On the Ground at HPE Discover 2026

HPE Discover felt enormous. The Venetian was packed with sessions, hands-on demos, and one of the largest expo halls we've seen at an enterprise event. The production quality was equally impressive. Main stage presentations felt closer to a broadcast than a traditional conference keynote.

The Steve Aoki and Imagine Dragons concert at Allegiant Stadium was another reminder that the best events create opportunities for conversations outside the conference halls. Some of the best discussions happen after the sessions end.

By the end of the week, one thing was clear: HPE invested heavily in the attendee experience, and it showed. Kudos to the entire HPE marketing team for this masterpiece execution.

Imagine Dragons lit up the week!

// UPCOMING EVENTS

Space Is Limited

Bay Area Startups Collectively Secured $16B+ in June MTD

Silicon Valley startup funding was slow this week, but the June total still moved up above $16B pushed by three megadeals. SandboxAQ received a $500M CHIPS award from the Department of Commerce to accelerate the AI-driven discovery of critical semiconductor materials. Odyssey (world models) closed a $310M Series B and Ent (cybersecurity) came out of stealth with a $100M seed.

This week also saw the largest-ever acquisition of a venture-backed company when SpaceX exercised their option to acquire Cursor (aka Anysphere) for $60B. Cursor had previously rejected at least two offers from OpenAI.

For startups raising capital: Stay on top of who's raising, who's closing and who's investing with the Pulse of the Valley weekday newsletter. Founders get the newsletter, database and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Ent closed a $100M Seed, the intent-aware Workspace Security platform extending endpoint security into a real-time layer of prevention across human and AI-driven work.
XDOF closed a $70M Series A, the infrastructure partner for the world's most ambitious robotics builders.
Convey closed a $38M Series A, the AI workforce platform that enables operators to train, deploy, and manage enterprise-grade digital teammates for modern businesses.
Architect Labs closed a $24M Seed, building an AI system that designs and verifies chips end-to-end.
GRU Space closed a $.1M Pre-Seed, building permanent lunar infrastructure to make humanity interplanetary within our lifetime, starting with the first hotel on the Moon.

Growth Stage:

Odyssey closed a $310M Series B, pioneering general world models: causal, multimodal systems that learn to predict and interact with the world over long horizons.
AttoTude closed a $52M Series C, a next-generation interconnect platform for AI and hyperscale infrastructure to enable high-bandwidth, energy-efficient connectivity
Cellares closed a $50M Series D, providing global cell therapy development and manufacturing services through an Industry 4.0 approach to the mass manufacture of living drugs.
Bland AI closed a $50M Series C, a voice AI platform for deploying production-grade AI agents across phone, SMS, and chat.
Articul8 AI closed a $29M Series B, a generative AI (GenAI) enterprise software company focused on helping organizations solve the world’s toughest problems.

AT HPE discover, AMD showcased Helios, a rack-scale AI system built around Meta's 2025 OCP Open Rack design.

The numbers are substantial: 72 Instinct MI455X accelerators, 31 TB of HBM4 memory, 1.4 PB/s of aggregate memory bandwidth, and up to 2.9 FP4 exaFLOPS of inference performance. Helios pairs the accelerators with 6th Gen EPYC Venice CPUs and AMD Pensando networking.

Just as notable is the architecture. Helios is being offered as an OCP-aligned reference design for OEM and ODM partners rather than a closed stack. Volume deployments are already underway in 2026. TCS has signed on to co-develop sovereign AI factory designs in India.

Open architecture. Rack-scale systems. Production deployments. Worth watching.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Your growth team woke up to a briefing they didn't ask for.

Monday 7am. Three messages in #growth.

Stripe revenue by channel, Meta and Google spend reconciled against GA4, Klaviyo flow performance, Shopify AOV by source. Posted by Viktor at 6am.

The campaign brief he wrote sits in #campaigns. Brand monitoring scrape runs every six hours. Competitor pricing update lands every Friday.

Your media buyer, content lead, and CMO open Slack to the same prepared room. 3,000+ integrations including every ad platform, CDP, and CMS you run.

"Viktor is like the most capable all-round colleague you can imagine." Sam, CEO, Givr.

Get Started for Free

𓃓 West Texas Pushes to Become AI's Next Power Center?💰 Bay Area Startups Collectively Secured $14B+ in June MTD

Sat, 13 Jun 2026 15:00:00 +0000

🛠️ Live from Midland: West Texas Pushes to Become AI's Next Power Center

At this week's 2nd Annual Permian Power Conference in Midland, Texas, one idea kept surfacing across conversations on stage and off: West Texas is moving from energy stronghold to AI infrastructure frontier. The event, held June 8–9 at Horseshoe Arena, drew more than 800 registrants, 100+ sponsors, and 60+ speakers spanning power, energy, data centers, capital markets, and AI infrastructure. (the-power-connection.com)

The case for the region is straightforward. Conference organizers framed the Permian as a place where land, stranded gas, power generation, permitting know-how, and infrastructure partners are converging at exactly the moment AI developers are searching for faster paths to capacity. The conference website goes further, calling the Permian "the next frontier in America's AI infrastructure buildout," noting that seven GW-scale AI hyperscale data center campuses have been announced in the region in the past six months.

That backdrop gave added weight to one of the conference panels on scaling power and infrastructure for AI. The central message from the discussion: the old playbook is breaking down fast. In the panel conversation, speakers described a rapid progression from grid-only assumptions to bridge-to-grid models, then to behind-the-meter strategies, and now increasingly to private-grid thinking. In practical terms, developers are no longer assuming the utility interconnect will define the project. Increasingly, they are designing projects around speed to power first.

That shift is being forced by scale. Panelists described how expectations have changed from 12–24 MW phases to 500 MW-plus requests, a jump that is exposing every weak point in the system: utility processes, equipment availability, construction logistics, modular deployment, labor, and permitting. One example shared on stage captured the mismatch well: only about a dozen 500 MW single-load interconnects have been completed in the U.S. over the past 60 years. Another: one co-op's "mega large load" tariff starts at just 2 MW, underscoring how far current frameworks are from what AI buyers are now requesting.

The panel also highlighted a widening gap between Silicon Valley demand and infrastructure reality. On the buy side, AI companies are moving at software speed. On the build side, land, power, and construction still move on physical timelines. Speakers described a market in which newer AI infrastructure companies often want to control multiple layers at once — GPUs, sites, and power — but don't always have the capital, credit profile, or operational maturity to do it. Credit friction came up repeatedly, especially for smaller and mid-sized neo-clouds seeking gigawatt-scale contracts without strong balance sheets or external backing.

That is where West Texas believes it has an opening. What stood out in Midland was not just resource availability, but ecosystem density. This is exactly how the conference describes its role: bringing together the energy producers, equipment providers, landowners, consultants, fiber companies, permitting experts, and data center stakeholders who "make or break your site selection."

The overall mood at the conference was not hype. It was urgency. The AI buildout is already colliding with the limits of the traditional grid-centric model, and the market is now looking for regions that can respond with pragmatism and speed. Based on the discussion in Midland this week, West Texas intends to be one of them.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today

Bay Area Startups Collectively Secured $14B+ in June MTD

After a slow start, June funding got a bump from a $12B series B for Jeff Bezos' startup Prometheus, and multiple megadeals, including SonoThera and Genspark AI.

Exits, IPOs, Liquidity: The big story for the valley this week is the SpaceX $75B IPO. The company may not be local, but multiple local VCs invested and are now holding stakes worth much more. Founders Fund invested $620M, now worth $50B+; Andreessen Horowitz invested $750M, now worth 10B+ (the biggest return in its history); Sequoia invested $1B+, now worth more than $20B. 137 Ventures accumulated a 1%+ stake through multiple transactions - primary investment, tender offers, secondary sales - over 15 years, now worth $20B+.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Maneva closed a $27M Series A, revolutionizes manufacturing by offering a superhuman digital workforce tailored to the industry's unique needs.
SyntheticFi closed a $13M Seed, fintech platform that helps registered investment advisors deliver low-cost and tax-efficient portfolio-backed financing solutions to clients.
SWARM Engineering closed a $10.5M Series A, the decision intelligence company for agrifood and manufacturing.
Billables AI closed a $10M Series A, AI-powered timekeeping and operational intelligence platform built for law firms and professional services.
Fearn closed a $5.5M Seed, AI-native patent platform built for a first-to-file world.

Growth Stage:

Prometheus closed a $12B Series B, building an "artificial general engineer" to drastically accelerate the design and manufacturing of complex physical products.
SonoThera closed a $125M Series B, developing an ultrasound-mediated nonviral genetic medicine platform designed to deliver the next generation of redosable genetic medicines.
Genspark AI closed a $100M Series B, an all-in-one AI Workspace on a mission to autopilot the busywork for 1 billion+ knowledge workers.
PhoenixAI closed a $80M Series B, enables enterprises to quickly and easily grow their business with a real-time analytical engine.
Coram AI closed a $35M Series B, an AI-native physical security platform that helps organizations investigate incidents faster and coordinate response.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Help us make better ads

Did you recently see an ad for beehiiv in a newsletter? We’re running a short brand lift survey to understand what’s actually breaking through (and what’s not).

It takes about 20 seconds, the questions are super easy, and your feedback directly helps us improve how we show up in the newsletters you read and love.

If you’ve got a few moments, we’d really appreciate your insight.

Take the survey.

🛠️ Everything. All at Once. The Infrastructure Stack Is Converging 💰 Bay Area Startups Collectively Secured $79B+ in May

Sat, 06 Jun 2026 15:00:00 +0000

🛠️ Everything. All at Once. The Infrastructure Stack Is Converging

The clearest signal from AIS5 wasn't any single technology shift — it was how tightly coupled those shifts have become. Power constraints are reshaping hardware choices. Hardware density is forcing cooling and networking changes. And all of it is being stitched together by software. The data center isn't a collection of independent layers anymore. It's one system.

Power Sets the Constraints

AI demand is growing faster than new power infrastructure can be delivered. That's pushing operators toward stranded power — existing capacity that often sits underutilized — rather than waiting years for new generation and transmission projects.

But using that power effectively isn't just an electrical problem. Software orchestration now manages power envelopes dynamically, increasing the compute supported by existing infrastructure. And as the industry pushes toward gigawatt-scale AI campuses, power planning requires coordination across utilities, developers, operators, and governments.

The conversation has moved past adding capacity. It's about redesigning how capacity is delivered — and that redesign ripples into every other layer.

Networking, Cooling, and the Density Problem

Higher-density deployments driven by new silicon require liquid cooling to manage thermal loads. Co-packaged optics reduce networking overhead at the rack level. These aren't independent upgrades — they're responses to the same pressure. When you pack more compute into less space, everything around it has to adapt.

The result is that facility design, cooling architecture, and network topology are now downstream of hardware and power decisions in a way they weren't five years ago.

What This Means

No single layer can be planned in isolation anymore. Power shapes hardware selection. Hardware density drives cooling and networking. Software ties it all together. The organizations building AI infrastructure at scale aren't optimizing one layer at a time — they're designing the stack as a single, interdependent system.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today

Bay Area Startups Collectively Secured $79B+ in May

After closing the month with a $36B funding week, June got off to a quiet start with just under $2B in funding, and five megadeals, including Generalist AI, Tripo AI and ZutaCore. Despite that, 2026 startup investment continues to set new monthly records.

Following Alphabet(Google)'s $85B stock sale to fund their AI plans, Meta is now considering a stock sale to raise tens of billions to fund their AI capital expenditures, per the Financial Times. Meanwhile, to ensure adequate compute supply, Google will pay SpaceX $920M monthly for accesss to Nvidia GPUs. And Apollo and Blackstone copmpleted a $35B package for Anthropic to lease TPUs.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Collate Software closed a $95M Series A, uses AI to create and streamline paperwork for diagnostic, medical device, and drug development companies.
Town AI closed a $55M Series A, building tools that make it easy for anyone to create and use software in their everyday work.
Terra AI closed a $20M Series A, solves the world's most critical subsurface problems with cutting-edge artificial intelligence.
Trajectory closed a $15M Seed, the platform for continual learning, turning real product usage into AI that continuously improves.
Drafted closed a $1M Seed, designs your dream home instantly with AI.

Growth Stage:

Supabase closed a $500M Series F, the open source Postgres development platform and leader in agentic infrastructure.
NewLimit closed a $435M Series C, developing medicines to treat age-related diseases by reprogramming the epigenome, a new therapeutic mechanism to restore regenerative potential in aged and diseased cells.
Generalist AI closed a $400M Series B, an AI robotics company with the mission to make general-purpose robots a reality.
ZutaCore closed a $100M Series C, waterless, direct-to-chip liquid cooling, removes it at the source by turning liquid into vapor, keeping processors consistently cool.
Advanced Nanotherapies closed a $31M Series B, is developing a differentiated functionalized nanoparticle platform for local, dual-drug delivery during catheter-based interventions.

Founder & CEO of SemiAnalysis, Dylan Patel

AI infrastructure is becoming harder to model as compute demand, power constraints, and deployment economics shift simultaneously.

Training clusters are scaling. Inference workloads are multiplying. Hyperscalers continue investing heavily in AI infrastructure while power availability, GPU supply chains, and data center construction timelines become increasingly important variables.

For operators, investors, and infrastructure leaders, understanding where those pressures are building has become a competitive advantage. Small shifts in one part of the stack can create significant downstream impacts across deployment strategies, capital allocation, and long-term planning.

SemiAnalysis is an independent research and analysis firm focused on semiconductors, AI infrastructure, hyperscaler economics, GPU markets, data centers, and the systems powering modern AI.

What they deliver

They produce deep technical and economic research across the AI infrastructure stack. Their work covers GPU supply chains, hyperscaler capital expenditure, inference economics, networking, power infrastructure, cloud pricing models, and AI data center forecasting.

Their research is widely referenced by infrastructure operators, investors, semiconductor companies, cloud providers, and organizations seeking a deeper understanding of where AI infrastructure is heading.

Who they serve

Hyperscalers, infrastructure operators, semiconductor companies, investors, cloud providers, and teams making long-term decisions around AI deployment and infrastructure strategy.

As AI adoption accelerates, understanding the economics and physical realities underneath the stack is becoming increasingly important. Questions around power availability, infrastructure utilization, deployment costs, and capacity planning are now shaping strategic decisions across the industry

Explore SemiAnalysis to follow deeper research on the systems, markets, and economics shaping the future of AI infrastructure.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Someone just spent $236,000,000 on a painting. Here’s why it matters for your wallet.

Late last year, a Klimt sold for the highest price ever paid for modern art at auction.

An outlier sure, but it wasn't a fluke. U.S. auction sales grew 23.1% in 2025. The $1-5mm segment even grew 40.8% YoY.

Meanwhile, Apollo’s chief economist Torsten Slok said to expect ‘zero in return in the S&P 500 over the coming decade.’

So, that's where it gets interesting.

Each environment is unique, but after dot-com, post war and contemporary art grew about 24% annually for a decade. After 2008, about 11% for 12 years.

It’s also had near-zero correlation with the S&P 500 since ‘95.*

Now, Masterworks lets you invest in shares of artworks featuring legends like Banksy, Basquiat, and Picasso.

$1.3 billion invested across over 500 artworks.
28 sales to date.
Net annualized returns on sold works held 12 months+ like 14.6%, 17.6%, and 17.8%.

Shares can sell quickly, but my subscribers can skip the waitlist:

Skip the waitlist

^{*Investing involves risk. Past performance is not indicative of future returns. See important Reg A disclosures at}^{masterworks.com/cd}^.

⚠️Why Most AI Pilots Stall Before Production 💰 Bay Area Startups Collectively Secured $36B+ in May Week 4

Sat, 30 May 2026 15:00:00 +0000

⚠️Why Most AI Pilots Stall Before Production

A large percentage of AI pilots still never make it into production.

Usually, the model works. The demo lands well internally. A team proves that a workflow can be automated or accelerated, leadership gets interested, and momentum starts building around the idea that AI can create a meaningful operational advantage.

Then the rollout slows down.

Ownership becomes unclear. Nobody defined what success actually meant beyond proving the model could function. Budgeting never moved beyond experimentation. What started as an engineering initiative never connected cleanly into procurement, operations, legal, security, or the teams expected to support it long term.

That pattern is becoming common across enterprise AI deployments.

The organizations moving faster tend to approach AI differently from the start. Projects are treated more like operational systems than isolated technical experiments. There is usually a named executive sponsor, clear accountability around deployment, and some level of centralized oversight to keep adoption from fragmenting across teams.

Many are building dedicated groups around deployment and governance, whether formal AI Centers of Excellence or smaller cross-functional teams responsible for standards, rollout strategy, and measurement. The goal isn’t centralization for its own sake. It’s reducing duplication, preventing disconnected implementations, and creating a repeatable process for moving projects into production.

The infrastructure side creates another layer of difficulty.

Most enterprise environments still operate across disconnected systems, inconsistent datasets, and workflows that were never designed around AI-driven access patterns. That creates friction long before compute becomes the limiting factor. Models depend on stable retrieval, clean data flow, and systems capable of supporting continuous interaction between applications and inference layers. Without that foundation, adding more GPUs rarely fixes the underlying issue.

This is where a lot of enterprise deployments start breaking apart.

The technical side of the model may work perfectly, but the surrounding organization and infrastructure can’t support it cleanly once usage expands beyond a small pilot group. Integration becomes difficult. Measurement becomes inconsistent. Different teams begin deploying overlapping systems without shared standards or visibility into how they interact.

What enterprises increasingly need is operational structure around deployment itself. Governance, integration, orchestration, measurement, and organizational alignment are becoming just as important as model performance.

That changes where the opportunity sits for builders and vendors.

A growing number of companies are no longer searching for another standalone AI tool. They are trying to figure out how to move existing systems into production without creating more operational complexity than they remove.

The technology continues moving quickly. Most organizations don’t.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today

Bay Area Startups Collectively Secured $36B+ in May Week 4

The last week of the month closed with more than $36B in fundings, with most of that coming from the last $35B of Anthropic's $65B Series H at a post-money valuation of $965B. That number puts them ahead of rival OpenAI's most recent $852B valuation. In the valuation space, Corgi Insurance caught attention this week when they announced a $106M Series B1 round with $2.6B valuation – just three weeks after their Series B at a $1.3B valuation.

2026 YTD SV funding is now over $300B, and money continues to be invested at unprecedented levels.

LinkSV-WITI webinar coming up: for a deeper dive on SV fundraising, valuation trends and impacts on the startup and venture ecosystem, join us next Friday, June 5, at noon for the monthly LinkSV-WITI webinar, Tech, Talent & Investment Trends. Register and grab the link here.

For startups raising capital: Stay on top of who's raising, who's closing and who's investing with the Pulse of the Valley weekday newsletter. Founders get the newsletter, database and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Reactor closed a $59M Series A, the developer platform for real-time generative video.
Countable Labs closed a $26M Series A, transforming how biology is measured.
Tensormesh closed a $20M Seed, the leader in caching-accelerated inference optimization for enterprise AI.
Itera closed a $12M Seed, enabling engineers to test real hardware changes in seconds instead of weeks through breakthrough fluid circuit technology
Canyon Code closed a $5M Pre-Seed, gives enterprises the controls to optimize, manage and govern their multi-agentic apps at scale

Growth Stage:

Anthropic closed a $35B Series H, an AI safety and research company developing AI systems that are helpful, honest and harmless.
Cognition AI closed a $1B Series D, an applied AI lab building end-to-end software agents.
Corgi Insurance closed a $106M Series B, an AI-native, full-stack insurance carrier built for startups.
Triomics closed a $22M Series B, a generative AI platform purpose-built for oncology workflows.
Signos closed a $20M Series B, on a mission to eliminate obesity by igniting the metabolic health movement

Most AI content online is still focused on the surface layer. Product demos, chatbot conversations, and recycled commentary about adoption curves.

Meanwhile, the people building the infrastructure underneath AI are dealing with a completely different set of problems.

Power constraints. Rack density. GPU orchestration. Cooling limits. Inference economics. Data movement. Energy timelines. Multi-tenant infrastructure. The operational reality of deploying AI systems at scale.

That is the value of our Ignite YouTube library.

Get Tuned In Or Risk Missing Out

The channel now contains more than 293 videos, including 40+ videos from AI INFRA SUMMIT 5 and a growing archive of shorts, operator clips, fireside conversations, infrastructure discussions, and full-length talks from across the AI infrastructure ecosystem.

The library features founders, hyperscalers, GPU cloud operators, infrastructure providers, investors, and enterprise builders discussing the bottlenecks, tradeoffs, and deployment realities shaping the next era of AI.

This is less about “AI trends” and more about understanding the physical and operational systems now driving the market forward.

Explore the IgniteGTM YouTube library to hear directly from the teams building AI infrastructure in real time.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Where to Invest $100,000 Right Now, According to Experts

Investors face a dilemma. When the S&P 500 finished its worst quarter since 2022 last month, diversifiers like bonds and bitcoin fell too.

Even with the turnaround in mid-April, analysts at Goldman Sachs and Vanguard have projected low-single-digit annualized returns from 2024-2034.

Bloomberg asked where experts would personally invest $100,000 for their March monthly edition.

One answer that surfaced for a second time? Art.

It's what billionaires like Bezos and the Rockefellers have privately used to diversify for decades.

Why?

Appreciation. The ArtPrice100 Index outpaced the S&P 500 overall from 2000 to 2025
Low-correlation. The postwar contemporary segment has moved independently of traditional investments like stocks since ‘95.*
Resilience. A scarce, physical, and global asset class with decades of demonstrated demand.

Thanks to the world's premier art investing platform, now anyone can invest in works featuring legends like Banksy, Basquiat, and Picasso, without needing millions.

Shares in new offerings can sell quickly but...

Our subscribers skip their waitlist*

^{*According to Masterworks data. Investing involves risk. Past performance is not indicative of future returns. See important Reg A disclosures at}^{masterworks.com/cd}^.

📉 Collapsing Token Pricing 📈 Rising Costs 💵 Bay Area Startups Collectively Secured $35B+ in May Week 3

Sat, 23 May 2026 15:00:00 +0000

📉 Collapsing Token Pricing 📈 Rising Costs

💡 The price of a single AI token has collapsed. Hardware improvements, inference engine optimization, and architectural changes like disaggregated prefill/decode systems have pushed token costs down fast. API pricing keeps getting cheaper as open models and infrastructure providers fight for efficiency.

By itself, that should have lowered enterprise AI spend.
…BUT IT DIDN’T.

Enterprise AI bills are climbing because cheaper tokens unlocked far more usage. (Yes, Jevon’s Paradox effect) We moved past the phase where AI handled short chatbot conversations and lightweight prompts. Tokens are now tied directly to production workloads: code generation, autonomous agents, background processing, internal copilots, research loops, and operational automation running continuously inside the business.

A growing percentage of AI compute now happens at test time instead of training time. Companies are no longer paying for a static answer. They are paying for reasoning depth, iteration, verification, memory management, tool usage, and agent coordination happening live during inference.

A single enterprise task can trigger massive hidden token generation behind the scenes. Multi-agent systems may explore several solution paths, validate outputs, call external tools, retry failed actions, and run internal reasoning loops before returning a final response. One prompt can quietly generate millions of tokens worth of compute.

So even as the cost per million tokens drops, overall infrastructure spend keeps rising because workload volume is scaling faster than efficiency gains.

This is changing how enterprises evaluate ROI around AI infrastructure.

The question is no longer “What does a million tokens cost?”

The question is “How much labor, operational throughput, and decision-making capacity did those tokens replace?”

Once companies see measurable gains from agentic systems, they become willing to consume enormous amounts of inference. At that point, lower token pricing stops reducing spend and starts accelerating adoption.

View full panel conversation

The unit economics are collapsing.
The aggregate infrastructure demand is exploding.

That is why the era of the multi-million dollar AI inference bill is only getting started.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today

Bay Area Startups Collectively Secured $35B+ in May Week 3

The third week of the month closed with more than $35B in fundings, 98% of that came in eleven megadeals. The three largest were in the billions – Anthropic, $30B; Unnamed Anthropic JV, $1.5B and AMP PBC, $1.3B. Among the other ten megadeals were the $100M stakes that the U.S. government took in nine quantum computing companies, including four in Silicon Valley - two startups (PsiQuantum, Atom Computing), and two public companies (D-Wave Quantum, Rigetti Computing).

Follow LinkSV on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Unnamed Anthropic JV closed a $1.5B Series A,an AI-native enterprise services firm to help mid-size companies bring Claude into their core operations.
AMP PBC closed a $1.3B Series A, AMP Infra PBC provides pooled, automated infrastructure on the global AI grid.
Hark closed a $700M Series A, building the most advanced personal intelligence in the world.
Vital Signals closed a $15M Seed, a health innovation company addressing how people understand and manage blood pressure over time.
Origin Lab closed a $8M Seed, technology platform turning licensed game worlds into structured training data for world models and multimodal AI.

Growth Stage:

Anthropic closed a $30B Series unknown, is an AI safety and research company developing AI systems that are helpful, honest and harmless.
Decart closed a $300M Series B, our efficient systems-level AI infrastructure enables a tenfold improvement in both training and inferencing of the largest generative models.
Exa closed a $250M Series C, developing novel representation learning techniques and crawling infrastructure so that LLMs can intelligently find relevant information.
Armada closed a $230M Series B, is the hyperscaler for the edge, delivering modular AI infrastructure from first deployment to AI factory.
Commure closed a $70M Series F, delivers next-generation AI infrastructure for health systems.

VAST DATA // AI infrastructure is running into a data problem at the same time compute demand is exploding.

Training clusters are scaling, inference workloads are multiplying, and agentic systems are increasing pressure on storage, retrieval, orchestration, and real-time data movement.

Most organizations are not looking for “more storage.” They want infrastructure that can feed GPUs efficiently, eliminate bottlenecks, reduce latency, and support AI systems operating at production scale. That is where VAST Data fits.

Who they are
VAST Data is an AI infrastructure company building a unified software platform that combines storage, database, compute orchestration, and real-time data services into a single AI operating system.

What they deliver
They provide AI-native data infrastructure designed for large-scale training, inference, analytics, and agentic workloads. Their platform unifies structured and unstructured data, multi-protocol storage, real-time data access, and distributed compute orchestration into a shared architecture built for AI-era workloads.

Who they serve
Cloud providers, hyperscalers, enterprises, research organizations, and AI infrastructure operators managing large-scale GPU environments and data-intensive workloads.

VAST Data joined us at AI Infra Summit 5, where conversations around inference economics, data movement, GPU utilization, and agentic infrastructure repeatedly surfaced across the event. As AI infrastructure shifts toward distributed, real-time, and increasingly autonomous systems, the data layer is becoming one of the most important parts of the stack.

Vaughn Stewart, VP of Systems Engineering of VAST Data talking on a panel at AIS5

Explore VAST Data to see how organizations are building AI infrastructure around unified data architecture.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Crash Expert: “This Looks Like 1929” → 71,105 Diversifying Here

Mark Spitznagel, who made $1B in a single day during the 2015 flash crash, warned markets are mimicking 1929. Seems extreme but we did just see the worst quarter for the S&P since 2022.

So it’s not so surprising that Vanguard and Goldman Sachs forecasted 5% and 3% annual S&P returns respectively for 2024-2034.

Late last year, Apollo’s chief economist Torsten Slok put it this way: "expect zero in return in the S&P 500 over the coming decade."

So, what’s something investors can actually do to diversify this week?

Almost no one knows this, but postwar and contemporary art appreciated 10.2% annually with near-zero correlation to equities from 1995–2025 overall.*

And sure… billionaires like Bezos can make headlines at auction, but what about the rest of us?

Masterworks makes it possible to invest in legendary artworks by Banksy, Basquiat, Picasso, and more – without spending millions.

29 exits. Net annualized returns like 16.5%, 17.6%, and 17.8% on works held over 1 year+. $1.3 billion invested. 500+ offerings.*

Shares in new offerings can sell quickly but…

👉 My subscribers skip the waitlist.*

^{*According to Masterworks data. Past performance is not indicative of future returns. Investing involves risk. Important Reg A disclosures:}^{masterworks.com/cd}^.

💡 When the Lights Flicker: AI's Real Bottleneck 💵 Bay Area Startups Collectively Secured $1.9B in May Week 2

Sat, 16 May 2026 15:00:00 +0000

💡 When the Lights Flicker: AI's Real Bottleneck

The pressure underneath AI infrastructure has become difficult to ignore. Compute keeps scaling, model demand keeps rising, and the physical systems underneath it are being pushed into conditions they were never designed to handle.

That tension showed up repeatedly at AI INFRA SUMMIT 5, particularly around one issue: power.

Compute Is No Longer The Constraint

The industry spent years optimizing around compute availability. Now the limiting factor is whether enough electricity can be delivered to the right place, at the right density, on the right timeline.

You can see it at the rack level first. Traditional facilities were built around workloads that look modest compared to modern AI systems. Higher-density deployments are pushing cooling, electrical routing, and facility design beyond the assumptions most existing infrastructure was built around.

Retrofitting Stops Making Economic Sense

That creates a difficult position for operators with older footprints. Some facilities can support incremental upgrades, but large-scale AI deployments often require major redesigns. In many cases, retrofitting stops making economic sense once the gap between legacy infrastructure and AI demand becomes too large.

Power timelines increasingly move slower than AI timelines. In major markets, grid interconnection delays stretch across years while demand for compute moves almost immediately. Capacity gets claimed early, projects move forward before power is fully secured, and operators are pushed toward whatever infrastructure can come online fastest.

Source: International Energy Agency, “Electricity 2025: Analysis and forecast to 2027”

Stranded Power Becomes Strategic

That pressure is driving more attention toward stranded and underutilized power. One of the more interesting discussions at AIS5 focused on how unused grid capacity could support AI infrastructure far sooner than waiting for entirely new power buildouts. Instead of sitting in interconnection queues for years, operators look for existing capacity and build around it.

The challenge is that AI workloads run continuously while grid conditions shift constantly around them. Weather events, regional demand spikes, and infrastructure stress all affect how much power can realistically stay allocated at any given moment.

Operators are starting to think about infrastructure that can adapt dynamically to available power conditions, scaling workloads up or down depending on what the surrounding system can support in real time. Once energy availability becomes variable, orchestration starts carrying much more weight across the stack.

The advantage increasingly goes to the teams that can turn available power into usable compute efficiently under constraint, not simply the teams deploying the most hardware.

The buildout continues, but the assumptions underneath it are changing fast. AI infrastructure now expands inside a system where power access, cooling capacity, and grid responsiveness all shape what can realistically scale.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today!

// UPCOMING

Tuesday, May 19 NYC

Bay Area Startups Collectively Secured $1.9B in May Week 2

The second week of the month closed with $1.9B in fundings, 74% of that in just four megadeals. The largest was a $650M Series A to Recursive Superintelligence, an AI research lab.

The Cerebras IPO beat expectations, pricing at $185 and opening the first day of trading at $350, valuing the chipmaker at $100B+ in the largest US tech IPO since Uber's in 2019. This may finally break the ice for a long line of late stage tech companies who have avoided going public since the end of 2022, preferring to close additional rounds of capital. Databricks is the poster child for this group, having closed a $5B Series L and tender offer for their employees in January.

For startups raising capital: Stay on top of who's raising, who's closing and who's investing with the Pulse of the Valley weekday newsletter. Click through to get more detail on investors and executives, including email addresses for both. Founders get the newsletter, database and alerts for just $7/month ($50 value). Check it out and sign up here.

Follow us on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

Recursive Superintelligence closed a $650M Series A, an AI research lab focused on AI that recursively improves itself.
Gridcare closed a $64M Series A, uses physics-based AI to identify and activate near-term capacity on today's grid.
Judgment Labs closed a $32M Series A, the platform for improving agents from production data.
Config closed a $27M Seed, building the data infrastructure that enables general-purpose bimanipulation.
Wirestock closed a $23M Series A, supplies datasets of images, videos, design assets, gaming and 3D content to AI labs.

Growth Stage:

Mind Robotics closed a $400M Series B, building the full-stack platform of foundation models, purpose-built robotics, and deployment infrastructure.
Cowboy Space Corporation closed a $275M Series B, our satellites will collect sunlight, convert it to electricity, and transmit power through infrared lasers to ground stations.
Exaforce closed a $125M Series B, our AI-native platform combines a real-time security knowledge graph with AI agents.
Cala Health closed a $50M Series D, our wearable neuromodulation therapies deliver individualized peripheral nerve stimulation.
Vapi closed a $50M Series B, lets enterprises deploy human-like voice agents in minutes.

Most companies are sitting on the same problem. Their data lives across disconnected SaaS tools, reporting systems, ad platforms, CRMs, spreadsheets, and operational dashboards. The result is slow decision-making, conflicting numbers, broken attribution, and AI initiatives that never make it past the prototype phase.

Teams are not looking for “more dashboards.” They want clearer visibility into the business, faster decisions, cleaner workflows, and AI systems that actually work in production. That is where Latticework Insights fits.

Who they are
Latticework Insights is an AI and data infrastructure firm focused on building custom AI stacks, analytics systems, and modern data platforms for consumer and retail brands.

What they deliver

They build custom AI agents, workflow automation systems, centralized data infrastructure, predictive analytics models, and operational reporting environments designed to unify fragmented business data into usable systems.

Zeitgeist of Big Data x AI Panel at AIS5 w/ Tim Shea, Nicole Wemple, Meenal Iyer & Okhtay Azarmanesh

Who they serve
Retail, ecommerce, and DTC brands operating across fragmented SaaS and analytics environments.

Latticework Insights joined us at AI Infra Summit 5, bringing a perspective grounded in what happens after the infrastructure is deployed. As more companies race to operationalize AI, the bottleneck is increasingly shifting toward orchestration, data quality, workflow integration, and production-ready systems that teams can actually trust.

Explore Latticework Insights to see how brands are building AI and analytics systems that move beyond demos and into real operational use.

CEO Tim Shea with Bill Barry

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Where to Invest $100,000 Right Now, According to Experts

Investors face a dilemma. When the S&P 500 finished its worst quarter since 2022 last month, diversifiers like bonds and bitcoin fell too.

Even with the turnaround in mid-April, analysts at Goldman Sachs and Vanguard have projected low-single-digit annualized returns from 2024-2034.

Bloomberg asked where experts would personally invest $100,000 for their March monthly edition.

One answer that surfaced for a second time? Art.

It's what billionaires like Bezos and the Rockefellers have privately used to diversify for decades.

Why?

Appreciation. The ArtPrice100 Index outpaced the S&P 500 overall from 2000 to 2025
Low-correlation. The postwar contemporary segment has moved independently of traditional investments like stocks since ‘95.*
Resilience. A scarce, physical, and global asset class with decades of demonstrated demand.

Thanks to the world's premier art investing platform, now anyone can invest in works featuring legends like Banksy, Basquiat, and Picasso, without needing millions.

Shares in new offerings can sell quickly but...

Our subscribers skip their waitlist*

^{*According to Masterworks data. Investing involves risk. Past performance is not indicative of future returns. See important Reg A disclosures at}^{masterworks.com/cd}^.

🌐 Internet's Next Phase: Web 4.0 👯‍♀️ Still feeling AIS5 Buzz 💵 Bay Area Startups Collectively Secured $6.4B in May

Sat, 09 May 2026 15:00:00 +0000

🌐 Internet's Next Phase: Web 4.0

Last week's AIS5 recap covered five signals that came out of our summit. Today’s will be more of a deep-dive, unpacking what "The Agentic Internet" actually means for the infrastructure stack and why it showed up as an active theme all day.

Jeremiah Owyang of Blitzscaling Ventures opened our lineup of speakers at AI INFRA Summit 5 by naming what he sees as the current phase: Web 4.0, the Agentic Internet. Autonomous AI agents are now deciding, transacting, learning, and replicating with minimal human input. The companies building ahead of this shift are producing 10X revenue per employee relative to traditional competitors.

Jeremiah Owyang, GP at Blitzscaling Ventures

That number got attention. But the part that stuck with the infrastructure crowd was what comes after the slide deck. If agents are making decisions, calling APIs, moving data, and triggering workflows on their own, the compute pattern underneath looks nothing like what most systems were designed for.

Traditional AI workloads follow a request-response pattern. A user sends a prompt, the model returns a result. Agentic workloads don't work that way. An agent receives a goal, breaks it into steps, calls tools, reads data, makes decisions, and sometimes spins up follow-on tasks without waiting for a human to approve each one. That changes the demand profile in a few important ways:

Compute becomes persistent. Agents don't spin up and shut down per query. They run continuously, maintaining state across long workflows. That turns burst demand into baseline load.
Token consumption compounds. Agentic workflows consume dramatically more tokens per task than single-turn interactions. Reasoning steps, tool calls, and multi-step chains all add up. This connects directly to Signal 3 from the AIS5 recap - inference economics are shifting because of how agents use the stack.
Security moves into the execution layer. When an agent can read files, make network calls, and access credentials on its own, the question of what it's allowed to do becomes an infrastructure problem. Sandboxing, policy enforcement, and audit trails have to be built into the runtime, not bolted on afterward.

Several AIS5 talks and sessions reinforced the same point from different angles. Nicola Tan from AMD laid out the case for rethinking enterprise infrastructure specifically around agentic AI. Luke Norris from Kamiwaza argued for AI orchestration as a core requirement of the agentic enterprise. The Token Economy panel, with representatives from Tencent Cloud and VAST, dug into what it actually costs when agents run continuously and consume tokens at scale. And in the breakout room, Michael Roy from AMD presented a session on why agents don't care what GPU they run on and what that means for teams designing heterogeneous compute environments.

Carmen Li, Vaughn Stewart and Wilson Liu

The common thread across all of them was the same. The infrastructure assumptions that work for single-turn AI break down when the workload is autonomous, persistent, and compounding.

Planning for agentic workloads means planning for compute that runs continuously, consumes tokens at a compounding rate, and requires governance built into the infrastructure itself. The organizations moving on this now are treating agents as persistent processes, not occasional tools.

Every organization sits somewhere on the AI culture spectrum. The ones building for agentic workflows are pulling ahead. The ones still treating AI as a feature are watching the gap widen.

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

See you at AIS 6 December 4th, San Francisco.
Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today!

AI INFRA 5 // STILL FEELING THE BUZZ

Edward Preston, Debbrah Dizon, Tim Shea, Maggie Zheng

See the video recap

Check out Photos from AIS5

// UPCOMING

Join The Webinar

Bay Area Startups Collectively Secured $6.4B in May Week 1

The first week of the month closed with $6.4B in fundings. 90% of the total was in seven megadeals, the largest being $4B to The Deployment Company (fka DeployCo.) That's the joint venture between OpenAI and several PE firms to sell AI tools to their portfolio companies. Anthropic has also announced a similar arrangement, and Google is reportedly in talks with other PE firms as well.

The Cerebras IPO will be the first in the AI arena in 2026, ahead of the SpaceX, Anthropic and OpenAI triad. And demand is high - the company plans to raise their IPO price range from the initial $115-$125 per share to $125-$135 after drawing orders for more than 20X the number of shares available.

Follow us on LinkedIn to stay on top of SV funding intelligence, and the companies, investors and executives impacting the startup ecosystem.

Early Stage:

RadixArk closed a $100M Seed, an AI infrastructure company building open, scalable systems for training, deploying, and running frontier models.
Standard Intelligence closed a $75M Series A, an aligned AGI lab building a general agent through full video pre-training on computer use.
Tessera Labs closed a $60M Series A, a multi-agent AI platform that transforms how enterprises modernize complex ERP systems and data.
Nace.AI closed a $21.5M Seed, an applied AI company building a specialized intelligence powered by Metamodel designed to run companies.
Altara closed a $7.6M Seed, building the scientific intelligence platform that helps frontier industries accelerate R&D through manufacturing.

Growth Stage:

Sierra Technologies closed a $950M Series E, helps businesses build better, more human customer experiences with AI.
Astranis closed a $450M Series E, builds advanced satellites for high orbits providing dedicated, secure networks.
Corgi Insurance closed a $160M Series B, an AI-native, full-stack insurance carrier built for startups.
Deep Infra closed a $107M Series B, a purpose-built cloud inference platform for high-throughput AI.
Anello Photonics closed a $25M Series B, develops integrated photonic system-on-chip technology.

vCluster is a Kubernetes infrastructure platform focused on virtual clusters, multi-tenancy, and GPU infrastructure orchestration for AI clouds and enterprise AI factories. The company helps operators deliver hyperscaler-style infrastructure experiences without multiplying physical clusters.

What they deliver

Virtual Kubernetes clusters with isolated control planes
Multi-tenant GPU infrastructure management
Self-service environments for AI and ML teams
Kubernetes orchestration across public and private cloud
Infrastructure automation for AI cloud providers and enterprise AI factories

CEO Lukas Gentele (left) moderating a panel at AIS5

Why it matters

AI infrastructure teams are running into the limits of traditional Kubernetes operations. Dedicated clusters for every team create overhead, waste GPU capacity, and slow deployment cycles. vCluster approaches the problem differently by virtualizing Kubernetes environments on shared infrastructure while preserving isolation, performance, and governance. The result is higher GPU utilization, faster provisioning, and infrastructure that scales more like a hyperscaler than a traditional enterprise environment.

Who they serve

AI cloud providers, platform engineering teams, enterprises operating internal AI factories, and infrastructure operators managing GPU-heavy workloads at scale.

vCluster joined us at AI Infra Summit 5, where conversations around Kubernetes orchestration, tenant isolation, GPU scheduling, and hyperscaler-grade infrastructure were front and center. As AI infrastructure moves from experimentation into industrial-scale deployment, the operational problems vCluster is focused on are quickly becoming core bottlenecks across the industry.

Explore vCluster to see how they are helping teams scale AI infrastructure more efficiently.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Apple’s Starlink Update Sparks Huge Earning Opportunity

Apple just secretly added Starlink satellite support to iPhones through iOS 18.3.

One of the biggest potential winners? Mode Mobile.

Mode’s EarnPhone already reaches 490M+ users that have earned over $1B, and that’s before global satellite coverage. With SpaceX eliminating "dead zones," Mode's earning technology can now reach billions more in unbanked and rural populations worldwide.

Their global expansion is perfectly timed, and investors like you still have a chance to invest in their pre-IPO offering at $0.52/share.

With their recent 32,481% revenue growth and newly reserved Nasdaq ticker, Mode is one step closer to a potential IPO.

Secure your shares in Mode Mobile at $0.52/share

^{Please read the offering circular and related risks at}^{invest.modemobile.com}^{. This is a paid advertisement for Mode Mobile’s Regulation A+ Offering.}

^{Mode Mobile recently received their ticker reservation with Nasdaq ($MODE), indicating an intent to IPO in the next 24 months. An intent to IPO is no guarantee that an actual IPO will occur.}

^{The Deloitte rankings are based on submitted applications and public company database research, with winners selected based on their fiscal-year revenue growth percentage over a three-year period.}

☁️ AI INFRA Summit 5: What 1,000 Infrastructure Leaders Heard ❤️ The Community Behind it all 💵 Bay Area Startups Collectively Secured $20.42B in April

Mon, 04 May 2026 13:35:29 +0000

AI INFRA Summit 5: What Infrastructure Leaders Heard

Luke Norris, Kamiwaza

The five signals that defined the day

On May 1st, AI INFRA Summit 5 crossed a threshold it hadn't reached before: 1,000 registered attendees gathered at Plug and Play Tech Center in Sunnyvale to hear from the builders, operators, and investors shaping the next wave of AI infrastructure. What emerged from a full day of keynotes, panels, and conversations wasn't a single narrative — it was five clear signals about where the industry is headed.

Signal 1: The Agentic Internet is already being built.

Jeremiah Owyang of Blitzscaling Ventures opened the day by naming where we are: Web 4.0, the Agentic Internet. Autonomous AI agents are now deciding, transacting, learning, and replicating with minimal human intervention. The companies building ahead of this shift are producing 10X revenue per employee relative to traditional competitors. The ones waiting are falling further behind. The question for every organization in the room was simple: where do you sit on the AI culture spectrum, and are you moving fast enough?

Jeremiah Owyang, GP at Blitzscaling Ventures

Signal 2: Power is the real bottleneck (and the biggest opportunity)

Across multiple sessions, one constraint dominated: power. Global data center power demand has tripled in three years. Rack density is climbing from 15kW toward 150kW and beyond. New grid interconnection timelines are measured in years, not months. Hammerhead's Bobby and Rahul Krur made the opportunity clear: the highest leverage play isn't waiting for new generation capacity — it's unlocking the stranded and underutilized power already inside existing infrastructure. Orchestration software that maximizes AI output within a constrained power envelope is a category that didn't exist five years ago and is now critical.

Dean Nelson, JP Balajadia and Rajesh Gopinath

Signal 3: Inference economics have changed the business model.

Training was last year's story. Inference is this year's constraint. Joo Young Kim of HyperAccel, Vasanth Mohan of SambaNova, and the Tokenomics panel all converged on the same point: per-token pricing is falling, but total AI costs are rising — because reasoning, agentic workflows, and long-context usage consume dramatically more tokens than previous generations. Token efficiency is now a design constraint, not a pricing footnote. The CFO is paying attention. The infrastructure stack is being re-architected around heterogeneous chips, distributed inference, and workload-specific optimization.

Signal 4: The enterprise AI failure mode is organizational, not technical.

Minal Iyer of SurveyMonkey delivered the most practically useful point of the day: 70% of AI pilots fail not because the tools don't work, but because there's no operating model behind them. No ownership. No governance. No repeatable measurement. The companies succeeding at enterprise AI scale aren't the ones with the best models — they're the ones with executive mandate, formal budgeting, and an AI Center of Excellence with a real ROI review cadence. Nicola Tan of AMD reinforced this from the infrastructure side: the production bottleneck is fragmented enterprise data, not model availability.

Signal 5: The infrastructure stack is being redesigned at every layer simultaneously.

Networking, power distribution, cooling, data orchestration, and chip architecture are all in transition at the same time. Co-Packaged Optics, liquid cooling, DC power architectures, distributed inference, and agentic orchestration aren't isolated innovations — they're converging on a new baseline for what AI infrastructure looks like. The companies that build around integrated, full-stack outcomes will set the standard.

The through line across every session was the same: the model layer is increasingly commoditized. The advantage belongs to teams who solve what's underneath it — power, data access, inference efficiency, and organizational readiness.

That's what AI INFRA exists to support. See you at Summit 6 — December 4th, San Francisco. Secure your spot with Super Early Bird Tickets below

Reserve Your Spot Today!

More Than a Conference: The Community Behind AI INFRA

Numbers tell part of the story. 1,000 registered attendees. Speakers from AMD, Supermicro, IBM Cloud, Tencent Cloud, Oracle, SambaNova, and a dozen more. A BMW i7 in the Ignite Studio. A main stage that ran from 9 AM to 5 PM without a quiet moment.

But the part that doesn't show up in the metrics is the reason people keep coming back.

AI INFRA Summit 5 was the fifth time this community has gathered — and the energy in the room felt different from a typical industry event. These aren't people attending to collect business cards. They're builders, operators, and investors who are actually trying to solve the same problems together. The conversations that mattered most on Friday didn't happen on stage. They happened in the Ignite Lounge between sessions, at the sponsor tables during lunch, and in the boardroom where capital partners met face to face for the first time.

That's what the AI infrastructure community looks like right now. It's a group of people who understand that the work being done — building the power, compute, networking, and data layers that AI actually runs on — is generational. The friendships formed here become the partnerships that close deals, solve hard problems, and build companies.

Bill Barry & Robert Scoble

We started AI INFRA as a place where serious infrastructure people could find each other. Five summits later, the community has built something that goes well beyond any single event. Thank you for being part of it. Summit 6 is December 4th in San Francisco. We'll see you there.

Check out Serena’s Vlog

Bay Area Startups Collectively Secured $20.42B in April

The last week of the month closed with $1.63B in fundings, taking April above the $20B mark, for the 4^th consecutive month of $20B+ in fundings. There were seventeen megadeals in April, including $500M for DeployCo., a joint venture between OpenAI and private equity firms TPG, Bain Capital, Advent International, Brookfield and Goanna Capital to sell AI tools to the PE firms' portfolio companies. The PE firms will be investing $4B into the joint venture.

Webinar Alert: LinkSV's Tech, Talent & Investment Trends is this Friday, May 8 @ noon PT. Join us for a look at April's $20B+ month to start Q2 and where that investment is going. We'll look at the silicon and software newcomers in AI infrastructure and what challenges + opportunities they bring to the startup ecosystem. Finally, the non-AI startups - which ones are finding investors and where? Register here for the link.

Early Stage:

Scout AI closed a $100M Series A, the frontier AI lab for war, developing the reasoning layer that turns commander intent into coordinated action at the edge.
OpenLight closed a $50M Series A, PASIC technology integrates all the components of silicon photonics devices, both active and passive components, into one chip.
ComfyUI closed a $30M Series A, a node-based workflow platform for generative AI, built for creators who require full control over their pipelines.
Lighthouse Pharmaceuticals closed a $12M Series A,a clinical-stage biopharmaceutical company developing novel small-molecule therapeutics for chronic neurodegenerative and inflammatory diseases.
General Analysis closed a $10M Seed, an AI security company that helps enterprises deploy agentic AI safely through adversarial evaluation and defensive tooling.

Growth Stage:

Musely closed a $360M non-dilutive capital, a prescription wellness company with a focus on accessible and effective telemedicine.
Hightouch closed a $150M Series D, a Composable CDP and Agentic Marketing Platform that empowers companies to power personalized marketing and business operations.
Netomi closed a $110M Series D, the enterprise agentic CX platform built for the world's most complex environments.
Parallel Web Systems closed a $100M Series B, the leading web search provider for AI agents, with APIs that give agents structured, grounded access to the open web,
MagicCube closed a $10M Series C, a device independent IoT security platform that protects against on-device, cloud, and network attacks.

Your Feedback Matters!

Logan Lemery
Head of Content // Team Ignite

Share the newsletter

Valley Recap

🪙 Nvidia’s Backstop Universe – Heads I Win, Tails Who Loses? 🌩️NEOCLOUD SUMMIT OCT8 💰 Bay Area Startups Collectively Secured $10B+ in September MTD

Nvidia’s Backstop Universe – Heads I Win, Tails Who Loses?

The $11T AI Buildout, Nvidia’s Backstop Economics, and the Limits of Nvidia’s Balance Sheet

An expanded view of Nvidia’s Balance Sheet

Why all the Guarantees and Backstops?

AI Cloud Agreements

Residual Value Guarantees under Capital Partnerships

Bay Area Startups Collectively Secured $10B+ in September MTD

Your Feedback Matters!

Logan Lemery Head of Content // Team Ignite

☁ NEOCLOUD SUMMIT: Where the Next Cloud gets BUILT 💰 Bay Area Startups Collectively Secured $3.7B in September Week 1 🇺🇸 Happy Labor Day

Where the Next Cloud Gets Built

This is where the next cloud gets built. And right now, nobody has it fully figured out.

Bay Area Startups Collectively Secured $3.7B in September Week 1

Your Feedback Matters!

Logan Lemery Head of Content // Team Ignite

Blu Dot surpasses 2,000% ROAS with self-serve CTV ads

THE PILOT IN THE MACHINE💰 Bay Area Startups Collectively Secured $17.7B+ in August MTD

We already made the case that AI strategy is really an org design problem. That argument stands. The next question is the one we actually argue about in the field, between meetings, on the drive home from a conference: what are the teams winning with AI doing differently, day to day?

Bay Area Startups Collectively Secured $17.7B+ in August MTD

Early Stage:

Growth Stage:

Tensor Machines

Your Feedback Matters!

Logan Lemery Head of Content // Team Ignite

Blu Dot surpasses 2,000% ROAS with self-serve CTV ads

YOUR AI INFRA CALENDAR FOR BACKHALF 2026 💰 Bay Area Startups Collectively Secured $16B+ in August MTD

Six months ago we started planning the busiest stretch IgniteGTM has ever run. Six events between September and December, each one chosen to build on the last.

Bay Area Startups Collectively Secured $16B+ in August MTD

Early Stage:

Growth Stage:

Your Feedback Matters!

Logan Lemery Head of Content // Team Ignite

Nobody actually knows how AI gets used.

Key Takeaways from Building the AI-Native Startup Stack 💰 Bay Area Startups Collectively Secured $10.7B+ in August MTD

The Next Advantage Isn't Model Access

Inside Tencent's Own AI Transformation

Voice AI Has Graduated

AI Media Is Shifting from "Generate" to "Direct"

Token Economics Are a CFO Conversation Now

The Bottom Line

Why This Audience and This Place Matters

Bay Area Startups Collectively Secured $10.7B+ in August MTD

Early Stage:

Growth Stage:

Axiado

Your Feedback Matters!

Logan Lemery Head of Content // Team Ignite

What's changing in global hiring?

TAKEAWAYS FROM THE SILLICON VALLEY AI TOUR💰 Bay Area Startups Collectively Secured $2.1B+ in August Week 1

TAKEAWAYS FROM THE SILICON VALLEY AI STUDY TOUR

Silicon Valley Isn't a Place. It's an Intersection.

Three Layers Moving at Three Speeds

Infrastructure Is the Bottleneck

The Iron Man Suit

What the Bankers Took Home

Bay Area Startups Collectively Secured $2.1B+ in August Week 1

Early Stage:

Growth Stage:

Helios

Your Feedback Matters!

Logan Lemery Head of Content // Team Ignite

Write docs 4x faster. Without hating every second.

WHY AI STRATEGY IS AN ORG DESIGN PROBLEM💰 Bay Area Startups Collectively Secured $14B+ in July MTD

Most of the conversation around AI and the workforce is trapped in two unhelpful extremes.

Bay Area Startups Collectively Secured $14B+ in July MTD

Early Stage:

Growth Stage:

Inflect

Your Feedback Matters!

Logan Lemery Head of Content // Team Ignite

The AI Agent You Can Trust

KEY TAKEAWAYS FROM AMD ADVANCING AI DAY💰 Bay Area Startups Collectively Secured $7.6B+ in July MTD

AMD just laid out its roadmap for the next era of AI compute and if you build, operate, or buy AI infrastructure, here's what matters.

1. The data center is the new unit of design

2. Helios is AMD's answer to rack-scale AI

3. The accelerator market is heading past $200B

4. The hardware roadmap extends through 2028

5. Software openness is the competitive wedge

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

Logan Lemery
Head of Content // Team Ignite

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

Logan Lemery
Head of Content // Team Ignite

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.

Logan Lemery
Head of Content // Team Ignite

The infrastructure underneath has to keep up.
That's what AI INFRA SUMMIT exists to support.