Ten-Year Bet on Cerebras: How a "Wafer-Scale AI Chip" Made It to Nasdaq

Original Title: Reflections on a decade with Cerebras

Original Author: Steve Vassallo

Original Translation: Peggy, BlockBeats

Editor's Note: On May 14, Cerebras officially listed on Nasdaq under the ticker CBRS. Its closing price on the first day represented an increase of approximately 68% from the offering price, making it one of the most anticipated AI hardware IPOs since 2026.

This article, written by Cerebras early investor Steve Vassallo, reflects on his nearly two-decade-long partnership with Andrew Feldman, spanning from SeaMicro to Cerebras. On the surface, it tells the venture capital story from term sheet to IPO, but it actually chronicles how a cutting-edge hardware company, during a period when consensus was unfavorable, bet on a fundamental restructuring of AI computing architecture. From wafer-scale chips and memory bandwidth bottlenecks to a series of engineering challenges including power delivery, cooling, and electrical continuity, Cerebras faced not just isolated technical hurdles, but the reinvention of an entire modern computing system.

What is most noteworthy is not that Cerebras ultimately produced a wafer-scale chip 58 times larger than traditional chips, but that the company chose a direction contrary to industry inertia from the very beginning: when GPUs had become the default answer for AI training, it sought to redefine what a computer built for AI should be. This required not only technical judgment and patient capital, but also a long-term, non-transactional trust relationship between investors and the founding team.

For today's AI hardware competition, Cerebras's significance lies in reminding the market that the computing revolution is not just about stacking more GPUs; it can also stem from reimagining the computing architecture itself.

The following is the original text:

On Friday, April 1, 2016, I emailed Andrew Feldman telling him I would climb over his backyard fence to personally deliver the term sheet for our investment in Cerebras.

It was April Fools' Day, but I wasn't joking.

Strictly speaking, this was not standard operating procedure for a venture capital firm. But by then, I had known Andrew for nine years and had been discussing his next company with him for nearly two. I wasn't going to miss out on this deal because of a few clauses being revised over a Saturday afternoon.

I first met Andrew in October 2007. He and Gary Lauterbach had just founded SeaMicro. I didn't invest in that round, but I connected with them deeply, particularly admiring their approach to thinking from first principles. I've kept an eye on them ever since.

Truly valuable relationships take time to cultivate. Truly valuable companies do too. Today, from the outside, Cerebras is a decade-old company about to go public. But to me, this is a nearly two-decade-long relationship that has finally reached the moment of ringing the bell.

August 2019: Andrew and me at the Hot Chips conference on Stanford campus. Cerebras launched its first-generation Wafer-Scale Engine there.

Deep Relationships and Unreasonable Ambition

When AMD acquired SeaMicro in 2012, I had a feeling Andrew wouldn't stay long in a large corporation. He possessed a fierce competitive spirit and a rebellious heart. By early 2014, he was already looking for an exit and we started meeting frequently, discussing what he might do next.

At that time, two things were far from being consensus: first, that AI would truly become useful; second, that the GPU was not the optimal computing architecture for AI.

Regarding the first point, many smart people I knew disagreed. After AlexNet emerged in 2012, some corners of the research community were already achieving near-magical results with convolutional neural networks. But in the broader software industry, AI still hovered somewhere between a marketing buzzword and a research project.

The second point, the hardware issue, had barely been raised seriously. The GPU had become the default choice for neural network training, primarily because researchers stumbled upon the fact that they were "less bad" compared to CPUs. Designing a new computing system specifically for AI workloads meant challenging the dominant architecture used by researchers worldwide.

But Andrew, Gary, and their co-founders Sean, Michael, and JP saw a different path. Each had accumulated decades of experience in chips and systems: Gary's background included pioneering work in dataflow and out-of-order execution in the 1980s; Sean focused on advanced server architecture; Michael handled software and compilers; JP specialized in hardware engineering. They were an exceptionally rare group: individually brilliant, but collectively possessing a multiplicative effect on their capabilities. They could envision an entirely new kind of computer.

They believed that if AI truly unlocked its potential, the resulting market size would far exceed the sum of all existing computing forms.

They also recognized the GPU's nature: it was originally a chip designed for graphics processing, temporarily promoted to an AI training tool on a new battlefield. While better than CPUs for parallel processing, no one would design a GPU-like architecture if starting from scratch for AI workloads. The real bottleneck limiting neural networks wasn't raw computational power, but memory bandwidth. This meant the chip they needed to create shouldn't optimize matrix multiplication within isolated cores, but rather how data flows efficiently through the entire computing structure.

Internally, investing in Cerebras was far from a consensus decision. Several of my partners had witnessed the previous semiconductor investment cycle ending mostly in losses and were very frank about their concerns. But ultimately, we agreed as a team. That weekend in April 2016, we told Andrew clearly: we wanted to be the first investor to give him a term sheet.

A few weeks later, Andrew, Gary, Sean, Michael, and JP moved into our EIR office space on the second floor at 250 Middlefield. I still have the floor plan drawn by our office manager. On that plan, Cerebras sat next to one of Foundation's founders, just a few doors away from Bhavin Shah, who later founded Moveworks. It was a good floor for a startup to grow.

Cerebras's first headquarters, on the second floor of our old office at 250 Middlefield.

Knowing Which Rules to Bend and Which to Break

Before Cerebras, the largest chip in computing history was about 840 square millimeters, roughly the size of a postage stamp. The chip Cerebras created was 46,000 square millimeters, 58 times larger.

Choosing wafer-scale chips meant accepting all the downstream design challenges that came with it. In nearly 80 years of computing history, no one had truly succeeded at this. This also meant no one had systematically solved these problems: How do you power such a massive chip? How do you cool it? How do you maintain electrical continuity across tens of thousands of connection points?

To achieve wafer-scale computing, Cerebras had to reinvent nearly every aspect of modern computing simultaneously: semiconductors, systems, data structures, software, and algorithms. Each direction alone could have been a startup. Andrew and his team chose to tackle the most difficult technical problems first. With their intense, almost tireless effort, they pushed through these problems one by one.

We held board meetings every six to eight weeks. They would update us on their attempts since the last meeting: a new system design variation, a new power delivery scheme, or a thermal management adjustment. By repeatedly confronting systemic challenges from all angles, they developed a hard-won clarity of expression. They would explain what they thought went wrong and what they intended to try next.

We would ask questions, then dig deeper with the team, mobilizing the people, resources, and relationships needed to help them find new breakthroughs. Six to eight weeks later, when we met again, the story would repeat on another technical problem: another frontier to explore. Each solution would reveal the next problem that had to be solved.

Their first prototype wafer smoked on initial power-up. The team called it a "thermal event"—a term you use for a fire when you don't want to scare the board or the landlord.

I was constantly calculating the power consumption per square millimeter, partly out of curiosity, partly because the numbers seemed too high to be real. So we brought in engineers from Exponent, a failure analysis firm whose former name was literally "Failure Analysis." They confirmed the power numbers were indeed as audacious as they looked, and helped us think through a series of solutions that wouldn't violate the second law of thermodynamics. After all, that was one law Andrew was smart enough not to argue with.

Engineering discipline lies in knowing which rules can be broken, which can be bent, and which must be respected. Andrew and his team possessed a battle-tested intuition for this distinction. They knew when they were challenging convention—which is what they set out to do; and when they were challenging the laws of physics—which they did not.

When you are building on the frontier, failure is inevitable. The only way to navigate failure is through discipline, persistence, and most importantly, trust: trust in the mission, trust in each other, and trust in the idea that after the first prototype self-destructs, you will still come back to the lab the next morning and start the next iteration.

This kind of work has no transactional version. It only has a long-term version: staying in the room through incomplete solutions and patient explanations. That way, when it finally succeeds, you are there to witness it.

That moment came in August 2019. Andrew, Sean, and their team stood in the lab, watching a brand-new computer they had designed themselves run for the first time. To an outsider, it appeared to do nothing interesting. In Andrew's words, it was probably as exciting as watching paint dry. But the difference this time was: no bucket of "paint" had ever truly dried before. They stood there watching for 30 minutes, and then went back to work.

Who You Build With Matters Profoundly

Some people choose problems based on what they know they can solve. Andrew chooses problems based on what he believes is worth solving. Incremental iteration doesn't excite him; he wants 1000x leaps. From day one, he wanted to build Cerebras into a generational, one-of-a-kind company.

This drive is partly a matter of character. Andrew describes it as a "disorder" of computer architects—being haunted by an idea for decades. But to me, it's more broadly a founder's "disorder." He looks at a problem, first asking: Can I make something that leads to a step-change improvement? Then he asks: If I succeed, will anyone care? If the answer to both is yes, he invests the next decade of his life into it.

This drive also comes from his upbringing. Andrew grew up surrounded by geniuses as naturally as most kids watch TV. His father was a pioneering evolutionary biology professor who played weekly doubles tennis with six rotating people. Three of those six later won Nobel Prizes, and one won a Fields Medal.

According to Andrew, these giants would patiently explain their work in physics, mathematics, and molecular biology in language a child could understand. He thus formed a profound impression of what genuine intelligence looks like, while also understanding, as his mother said, that being smart doesn't mean you have to be an asshole.

I later realized this was one of Andrew's core traits, as important as his rebellious ambition and his almost phototropic instinct for problems truly worth solving. He was deeply convinced that the most extraordinary people he had met were also remarkably kind.

This belief shaped how his team came together to accomplish incredibly difficult things. The first 30 people Cerebras hired had all worked with him before; some had been with him since 1996. Today, Cerebras has about 700 employees, roughly 100 of whom have followed him across multiple companies.

August 2022: Cerebras founding team at the Computer History Museum. From left to right: Sean Lie, Gary Lauterbach, Michael James, JP Fricker, and Andrew Feldman.

Importantly, kindness and competitiveness are not mutually exclusive. Andrew is intensely driven to win. He likes to say he is a professional version of David, fighting Goliath. Goliath is slow and always guarding against a frontal assault, leaving space for every other approach. David's advantage lies in appearing in ways and places Goliath cannot.

At SeaMicro, Andrew's largest channel partner in Japan was NetOne. NetOne's main supplier was Cisco, which entertained partners with private jets and yachts worth more than most houses in Palo Alto. Andrew's budget was much humbler, so he invited NetOne's CEO to his backyard for a barbecue. Later, the CEO told him he had done business with Cisco for decades but had never been invited to anyone's home. This seemingly small, deeply human gesture—something Goliath would never think to do—cemented their relationship.

From the First Term Sheet to the IPO

This morning, Andrew rang the opening bell at Nasdaq. I stood by his side. It has been ten years and 2,600 miles since it all began in our 250 Middlefield office.

Today, there are still rare founders doing what Andrew did back then: drawing on whiteboards at 3 AM, wrestling with unsolved technical problems. They, too, possess a fierce competitive spirit and a rebellious heart. They are trying to find a true partner willing to fight alongside them: someone who will dig into problems with them when the first prototype won't power up, and who will stay until it finally runs.

These are the founders I want to support: those who choose problems worth solving, imagine solutions 1000 times better than the status quo, and continuously refine and persevere through the inevitable challenges along the way.

For founders like Andrew, Gary, Sean, Michael, and JP, I would climb over a backyard fence on a Saturday afternoon to hand-deliver a term sheet.

Original Link