Etched - Building AI Hardware to Make Inference Faster and Cheaper
June 30, 2026
AI Summary
5 min readGavin Uberti and Rob Lockett founded Etched in 2023 to build a chip designed specifically for AI inference, not training. Their first product, a complete rack system, is already taped out and has over a billion dollars in customer demand. The company raised $800 million to build it. The bet is that inference — running models after they are trained — will become one of the largest markets in the world, and that the hardware built for it must be designed from scratch, not retrofitted from GPUs built before ChatGPT existed.
The technical bets: low voltage and cluster-scale memory
Etched’s architecture rests on two core technical bets. The first is what they call low voltage inference. The key insight is that voltage is quadratically proportional to power: halving voltage cuts power by a factor of four. GPUs today run at relatively high voltages, which means adding more flops to a chip causes thermal throttling — the chip self-regulates and slows down to avoid overheating. Etched designed a new power delivery mechanism that runs at under half the voltage of any other AI chip, allowing them to cram far more flops into the same silicon area without thermal limits. The second bet is on cluster-scale memory. For the decode phase of inference — generating output tokens — the bottleneck is memory bandwidth. But the relevant metric is not bandwidth on a single chip; it is bandwidth a
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of Invest Like the Best with Patrick O'Shaughnessy
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 (00:28) **Introduction and Early Skepticism** - Patrick introduces Gavin Uberti and Rob Lockett, founders of Etched, and the episode's framing: building a better AI chip against all odds.
- 2 (03:26) **The Technical Bets: Low-Voltage Inference and Cluster-Scale Memory** - Gavin explains the two core technical bets that differentiate Etched's architecture for inference.
- 3 (08:31) **Why Inference is the Bottleneck** - The founders argue that inference will become the largest market in the world, and that current hardware was designed before ChatGPT.
- 4 (10:20) **Founder Stories: Rob's Cancer and Gavin's Kernel Work** - Personal motivations that led to founding Etched.
- 5 (15:43) **The Bimodal Talent Philosophy: Legends and Chips on Shoulders** - How Etched recruits and pairs elite veterans with raw, first-principles young talent.
- 6 (24:34) **Velocity, Vertical Integration, and Parallelization** - The operating principles that let Etched move faster than incumbents.
- 7 (31:47) **The Hardest Technical Problem: Aligning Clock Signals to 50 Picoseconds** - A story of solving an "impossible" chip bug.
+ Full timestamped outline available in the app
Show Notes
My guests today are Gavin Uberti and Rob Wachen, the founders of Etched.
A few years ago, when they set out to build a better AI chip than the largest companies in the world, almost everyone I called told me it could not be done. They have since done it, taping out a working chip on their first attempt and becoming the first hardware company founded after ChatGPT to do so.
They already have more than a billion dollars of customer demand for their first product, and have raised eight hundred million dollars to build it.
Etched builds chips and systems designed to run AI models faster and at lower cost. They started the company in 2023, and that product is a complete rack for inference, the chip along with the boards, the power delivery, the interconnects, and the manufacturing to produce it all.
We talk about the technical bets behind their architecture, how they hired industry legends and paired them with elite 22 year-olds, and why they believe inference will become one of the largest markets in the world.
I think you will find the story of what they have built hard to forget.
Please enjoy my conversation with Gavin and Rob.
For the full show notes, transcript, and links to mentioned content, check out the episode page here.
-----
Become a Colossus member to get our quarterly print magazine and private audio experience, including exclusive profiles and early access to select episodes. Subscribe at colossus.com/subscribe.
-----
Ramp’s mission is to help companies manage their spend in a way that reduces expenses and frees up time for teams to work on more valuable projects. Go to ramp.com/invest to sign up for free and get a $250 welcome bonus.
-----
Trusted by thousands of businesses, Vanta continuously monitors your security posture and streamlines audits so you can win enterprise deals and build customer trust without the traditional overhead. Invest Like the Best listeners get a special offer of $1,000 off Vanta when you go to vanta.com/invest.
-----
WorkOS is the infrastructure B2B and AI-native compan
More from this podcast
Invest Like the Best with Patrick O'Shaughnessy →