Brevis

The Infinite Compute Layer for Web3 and Everything. ZK Power, Real Adoption.

[Brevis @brevis_zk]

say it again, slower. 🥶

[Brevis @brevis_zk]

In AI coding, generating the code is the easy part now. The hard part is telling whether what you generated is actually better. Anyone can get a model to rewrite a function; proving that rewrite is faster and still correct takes real work.

That's the part AutoEng handles. It measures every change against the real workload and keeps only the ones that hold up. Learn more 👇

[Brevis @brevis_zk]

Today we're introducing AutoEng: an AI agent that runs the entire performance-engineering loop on its own. Find the bottleneck, write the fix, prove it's correct, prove it's faster, keep only what wins.

We pointed it at Pico, our zkVM, and with no human in the loop it found a ~11% speedup. 🧵

[Brevis @brevis_zk]

keep it moving 🏃‍♂️🟠 gm.

[Brevis @brevis_zk]

We built AutoEng to optimize code on its own, no human in the loop. The whole thing rests on one design choice: it can't fake a win.

❌It can't cherry-pick.
❌It can't call a change "faster" without proving it byte-for-byte and beating a benchmark.
❌It can't talk itself into a good result.

Most "AI writes code" tools can do all three.

How it works ↓
https://t.co/WIgs5YbN9r

[Brevis @brevis_zk]

Gm, do you see what I see?

[michael @no89thkey]

For anyone wondering what on-prem proving for Ethereum looks like: we took a picture of our server before spinning this up!

[Brevis @brevis_zk]

Zkat's celebrating our newest AutoEng launch, come join! 🎉

[Brevis @brevis_zk]

Our on-prem @ethereum ZK Proof cluster is live. Proving on self-owned hardware is how Ethereum stays decentralized without centralized cloud chokepoint and how real-time ZK proving scales. Very proud to be the first team up-and-running in this initiative, with a generous grant from the Ethereum Foundation.

[Ethproofs @eth_proofs]

Congrats to @brevis_zk for being the first team to get up-and-running w/ their on-prem cluster for the Ethproofs On-Prem Proving Initiative!

https://t.co/fBaHqFknkx

[Brevis @brevis_zk]

This is what autonomous performance engineering looks like. 🧑‍🔧

It profiles the live system, proposes a change, and keeps it only if measurement proves it correct AND faster.

No human in the loop. No guesswork. No speedup it can't prove.

That's AutoEng, learn more:
https://t.co/WIgs5YbN9r

[Brevis @brevis_zk]

We zip the chaos away, but the proof stays. 🟠

[Brevis @brevis_zk]

Today we're introducing AutoEng: an AI agent that runs the entire performance-engineering loop on its own. Find the bottleneck, write the fix, prove it's correct, prove it's faster, keep only what wins.

We pointed it at Pico, our zkVM, and with no human in the loop it found a ~11% speedup. 🧵

[Brevis @brevis_zk]

Most "AI writes code" stories stop at generation and leave a human to check the result. AutoEng automates the judging too, and it can't be talked into a bad result.

A small but real glimpse of autonomous performance engineering.

Full write-up 👇
https://t.co/4VTyMVwcfV

[Brevis @brevis_zk]

Gm. some things are built to last. 🗿

[Brevis @brevis_zk]

gbrevis 🎇 good food, good people, and math you can trust. happy 4th!