Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Writing formal specifications has always been the bottleneck in formal verification. Today we're launching AutoProver, an agentic formal verification system that reads your code, generates specifications, and verifies them. 🧵

44,102 Aufrufe • vor 1 Monat •via X (Twitter)

32 Kommentare

Profilbild von Certora
Certoravor 1 Monat

AutoProver starts with your repository, using documentation and design docs to better understand your system. From there, it generates formal specifications describing how your code is supposed to behave.

Profilbild von Certora
Certoravor 1 Monat

Those specifications become tests and formal verification rules. You can review, commit, and maintain those assets alongside your codebase.

Profilbild von Certora
Certoravor 1 Monat

AutoProver runs the generated tests and proofs, investigates failures, and explains the results. It also checks for bugs outside the inferred specifications, drawing on patterns learned from years of formal verification and audit work.

Profilbild von Certora
Certoravor 1 Monat

After each run, you get a report with: • Implementation bugs • Potential design issues • The status of every generated property, test, and proof You can review the results and provide feedback, which will improve your future runs.

Profilbild von Certora
Certoravor 1 Monat

Software development is changing quickly and formal verification needs to keep up. AutoProver makes formal verification accessible to developers who aren't experts in it.

Profilbild von Certora
Certoravor 1 Monat

AutoProver Beta is live today for Solidity, with Rust coming soon. Try it at: Read more at:

Profilbild von Antonio Viggiano
Antonio Viggianovor 1 Monat

Congrats on the launch

Profilbild von Certora
Certoravor 1 Monat

thanks! 🫡

Profilbild von Antonio Viggiano
Antonio Viggianovor 1 Monat

- github integration is not working - I think AutoProver should be selected by default. This is the hot new stuff, it doesn't make sense that it is the 3rd option

Profilbild von Certora
Certoravor 1 Monat

Thanks for the feedback! GitHub integration has been fixed ✅

Profilbild von Ben Sparks
Ben Sparksvor 1 Monat

This looks so sick

Profilbild von Certora
Certoravor 1 Monat

🫡🫡

Profilbild von Radu | Linx
Radu | Linxvor 1 Monat

Looking forward to test it our with Rust when it’s available

Profilbild von Certora
Certoravor 1 Monat

coming very soon! stay tuned

Profilbild von Hubris
Hubrisvor 1 Monat

Massive, congrats to all the team!

Profilbild von Certora
Certoravor 1 Monat

Thanks 🙏

Profilbild von George Gorzhiyev
George Gorzhiyevvor 1 Monat

All the more reason I'm glad I have AI generate a ton of clean code/system documentation. That'll help AutoProver when I am ready to try it out :).

Profilbild von Certora
Certoravor 1 Monat

absolutely! let us know once you start using it, we're happy to help!

Profilbild von Ξlliot
Ξlliotvor 1 Monat

Huge win for the ecosystem! Looking forward to trying it out

Profilbild von Certora
Certoravor 1 Monat

Thanks! Happy to hear your feedback :)

Profilbild von Wyatt Benno
Wyatt Bennovor 1 Monat

This is cool :) what languages does it support?

Profilbild von Certora
Certoravor 1 Monat

thank you! it supports Solidity, Rust coming very soon

Profilbild von 256cisco
256ciscovor 1 Monat

Great release Certora Team

Profilbild von Certora
Certoravor 1 Monat

Thanks!

Profilbild von Savant.chat
Savant.chatvor 1 Monat

Does it flag when the inferred spec disagrees with the design docs?

Profilbild von Czar102
Czar102vor 1 Monat

Congrats to the team! Will definitely test it out and give feedback.

Profilbild von Athan Tsokolas
Athan Tsokolasvor 1 Monat

The hard part is not only generating the spec—it’s preserving the review boundary around it. A useful trail should show which code/docs informed a proposed invariant, what changed since then, and which human accepted it. Otherwise an agent can produce a formally valid answer to a stale model of the system.

Profilbild von Keags
Keagsvor 1 Monat

Definitely a good first approximation, but I worry that unless the proper care is taken in reviewing the spec this can produce false confidence. LLMs discharge proof obligations super well but the specification itself is still more art than science and requires careful judgement.

Profilbild von Pasty NFTs
Pasty NFTsvor 1 Monat

about time dev experience got saved

Profilbild von Crypto Jobs Hub
Crypto Jobs Hubvor 1 Monat

Are you guys hiring 👀

Profilbild von Certora
Certoravor 1 Monat

Yes:

Profilbild von Adel Bucetta
Adel Bucettavor 1 Monat

the honest answer is that a lot of developers have just accepted specs as a necessary evil, but auto-generated ones could be a game-changer for productivity and accuracy don't see this getting talked about enough

Ähnliche Videos

How do you actually formally verify the code underpinning Ethereum's future? In this episode (the finale of the lean Ethereum miniseries), Nico sits down with Alex Hicks (Alexander Hicks), lead of Protocol Snarkification at the Ethereum Foundation, to break down formal verification from first principles. They cover: – What formal verification actually is and the trust boundaries between proof assistants, SMT solvers, and kernels – The full verification stack for RISC-V ZKVMs: from SAIL specs to constraint extraction to soundness proofs – Why writing constraints directly in Lean makes proofs 10–100x more ergonomic – How AI is now proving hard theorems in hours for $200 — and what that unlocks for the whole pipeline They also explore the boundaries problem, why specs can have bugs too, and the end goal of a full Lean stack that bypasses Rust and LLVM entirely. Listen to the full episode ------------------------------------------------------------ TIMECODES: 09:16 – What is formal verification? Proof assistants vs SMT solvers 18:33 – Formal verification of code: specs, semantics, and trust boundaries 29:30 – Formally verifying the Lean Ethereum stack: RISC-V ZKVMs in focus 33:02 – Extracting ZKVM constraints into Lean and proving soundness 36:35 – Writing constraints directly in Lean: 10–100x better proof ergonomics 44:02 – Proving Polishchuk–Spielman in 8 hours for $200 with AI 51:01 – The end goal: a full Lean stack bypassing Rust and LLVM

Zero Knowledge Podcast

15,261 Aufrufe • vor 5 Monaten

There’s a popular theory that AI will finally make formal verification mainstream because mathematical proof of correctness will be needed when machines write most or all of the code. But will this happen? Hillel Wayne is one of the best people to answer. Timestamps: 00:00 Intro 04:32 The Crossover Project 11:37 What software engineering does better 15:30 What traditional engineering does better 18:17 Formal methods 29:32 TLA+: what it is and demo 36:58 TLA+ at Amazon 38:10 Ways distributed systems break 41:03 Formal methods and systems thinking 46:20 The value of learning math 50:23 What TLA+ is good for and isn’t 52:50 Alloy: a declarative language for software modeling 58:53 Other formal methods tools 1:01:24 Property-based testing 1:05:31 AI and the need for formal verification 1:12:29 Logic for programmers 1:14:35 Hillel’s 2025 prediction on AI’s impact 1:21:30 Book recommendation Brought to you by: • Antithesis – verify your system’s correctness without human review or traditional integration tests – and avoid bugs or outages. • turbopuffer – a vector and full-text search engine built on object storage. It’s fast, cheap, and extremely scalable. • WorkOS – everything you need to make your app enterprise ready. Two things I found especially interesting, talking with Hillel: 1. Amazon used TLA+ to find a bug almost impossible to locate without formal methods. In the paper How AWS uses formal methods, the AWS team shared that they’d found a complicated bug for which the shortest error trace to exhibit was 35 steps (!!). The bug passed unnoticed through extensive design review, code reviews, and testing. AWS concluded they wouldn’t have uncovered it if they’d stuck to conventional testing approaches. 2. Why not use formal verification for everything, then? It’s because specs in the real world are a nightmare to write. Even a simple problem like “find the file in a directory that has the most lines” gets complicated when modeled with formal methods. We would have to answer questions like: ‘do we look at ASCII or UTF-8 new line characters, what about unreadable files, and Symlinks?’ Without formal methods, we can write a simple verification that is right in 99%+ of cases. Formal methods require a lot of extra effort for the less than 1% of exotic use cases!

Gergely Orosz

34,678 Aufrufe • vor 1 Monat

Xavier Leroy (creator of OCaml) is an expert in compilers, formal verification of software and functional programming. This interview should be an approachable resource if you're curious about formal verification of software since I was learning that on the fly during it. In this episode: • OCaml compared with Rust and JavaScript • What is formal verification and how does it work • How languages call each other across boundaries • How to address "almost-correct" LLM code • How type inference works in programming languages Where to watch: • YouTube - • Spotify - • Apple Podcasts - • Transcript - Thank you to the sponsor of this episode for supporting my work: • WorkOS: makes your app Enterprise Ready with easy to use APIs to add SSO, SCIM, RBAC, and more in just a few lines of code, check them out at Chapters: 00:00 - Intro 00:43 - What sets OCaml apart 04:39 - OCaml vs Rust 07:57 - Why is manual memory management more performant 11:21 - Javascript vs OCaml 14:00 - Famous Rob Pike quote 16:05 - Type inference and how it works 22:12 - What is formal verification and how does it work 40:07 - What made multicore support difficult for OCaml 50:17 - How programming languages interface and call each other 57:41 - The danger of almost-correct LLM code 01:05:39 - How LLMs will change programming languages 01:10:26 - Industry vs academia 01:15:05 - Most interesting unsolved problems 01:18:30 - Top book recommendations for engineers 01:21:17 - Advice for his younger self 01:23:31 - Outro

Ryan Peterman

23,942 Aufrufe • vor 1 Monat