Unacceptable censorship in the Rust security community. @kaepora @RustSec
flyingpenguin.com/rustsec-integr…
English
Petar Maksimović | Nethermind
64 posts
Petar Maksimović | Nethermind
@PetarMax
Formal Verification Engineer and BD Lead (@NethermindSec). Web3, program analysis, photography, singing, tennis, mental health. All views are my own.
London, UK Katılım Ekim 2012
316 Takip Edilen236 Takipçiler
KEVM is an executable semantics. It does not "formally verify" the" entire" EVM in any way. What it actually does is reconcile the yellow paper with the official test suite and serves as a basis for formal verification of EVM bytecode through the Kontrol tool.
Wonder which team members who built KEVM would agree with this statement. As one such (former) member, I certainly don't.
English
Nethermind@Nethermind
Agentic systems are moving from demos to production. Most teams are focused on what agents can do. Almost nobody is focused on what the infrastructure underneath can sustain when agents operate at speed. Protocol performance becomes the constraint. When agents execute governance actions, manage capital, or coordinate across chains, worst-case throughput matters, not average case. Nethermind has been benchmarking this at the execution layer. Worst-case block execution under realistic state, across client implementations. The gap between what clients can handle and what protocols currently permit is significant. @marchhill1 on execution client benchmarking at the OP Scaling Summit, Tuesday. @icedcool_eth demoing our Aztec infrastructure tooling, Thursday. @PetarMax on formalizing FRI in Lean, Friday. Building systems where agents operate under real constraints? Our protocol engineering and security teams are in Denver all week.
English
Petar Maksimović | Nethermind retweetledi
Our Formal Verification team is beginning work on the @brevis_zk Pico RISC-V zkVM, building on our work on SP1 and OpenVM.
We are grateful to the @EthereumFndn for supporting this effort.
English
Presenting "Formalising FRI: A Lean Blueprint of Security" at ETH Denver, Friday Feb 20, 12:40 PM (Futurllama Stage).
The FRI protocol is a core component underpinning numerous STARK-based systems. No implementation has a formal proof of correctness. We have made an important first step in this direction by fully formalizing FRI in @leanprover, allowing implementations to be verified against this formalization in future.
This is what we do at @NethermindSec. We formalize and formally verify security properties of production-ready zkVMs (SP1 and OpenVM), zk verifiers (zkSync), and protocols (IntMax).
If you're designing protocols or building ZK infrastructure where you need the strongest correctness mathematical guarantees available on the market, come find me on Friday or anytime during the week!
English
It was a great pleasure to lead the formal verification of OpenVM. This is the most complete FV effort to date of a real-world RISC-V zkVM.
Big thank you to @theyisun and @jonathanpwang from @axiom_xyz for their support in this process, and to @ethereumfndn for the funding.
Nethermind Security@NethermindSec
Our Formal Verification team led by @PetarMax, with support from @EthereumFndn, has verified in Lean the correctness of the OpenVM RISC-V extension built by @axiom_xyz. This work proves instruction-level correctness and, for the first time, execution and memory consistency. 🧵
English
Petar Maksimović | Nethermind retweetledi
ZK circuits are defined by mathematical constraints over finite fields.
Their correctness is not a matter of coverage.
It is a matter of whether the equations enforce the intended semantics.
1/
English
Petar Maksimović | Nethermind retweetledi
The Nethermind Formal Verification team has completed its work on formalizing the FRI protocol in Lean and blueprinting its proof of soundness.
Supported by a grant from @ethereumfndn.
This result is historically unique.
🧵
English
@ilyasergey Yes, and what was the complexity of the proof? Did it have any creative insight or was it just brute force and case analysis? Does the paper cover this in more detail?
English
@PetarMax We gave it a math fact with an admitted proof we felt it would take us a few hours to produce. It returned a proof in a few minutes.
English
The proof of the last remaining Lean theorem in our upcoming conference submission has now been completed with the help of AI.
The research community’s perception of program verification is about to change irreversibly.
English
@pumatheuma @jtguibas @tamirhemo @dtumad @SuccinctLabs @JulekSU @NethermindSec Stay tuned for more exciting news next week, and do get in touch with if you have a zkVM that needs some formal verification!
4/
English
@pumatheuma @jtguibas @tamirhemo @dtumad @SuccinctLabs This, of course, would not have been possible without the amazing work of @JulekSU and the rest of the Formal Verification team at @NethermindSec.
3/
English
It was a privilege to lead this ground-breaking verification work from the @NethermindSec side.
We now have in our hands the first formally verified** RISC-V zkVM, and that is @SuccinctLabs's SP1 HyperCube, 64-bit edition no less.
** Terms and conditions apply, see blog.succinct.xyz/nethermind-lea….
1/
Nethermind Security@NethermindSec
We are delighted to announce that the @NethermindSec Formal Verification team, led by @PetarMax, in tandem with @SuccinctLabs, and with generous support from @ethereumfndn, has formally verified the correctness of all of the core RV-64 chips from the SP1 HyperCube RISC-V zkVM in @leanprover. 1/
English
English
@PetarMax @RosuGrigore @chenxiaohong @pisquared I wouldn’t worry too much about fake accounts unless they’re impersonating real people, spreading misinformation, or trying to scam others. Beyond that, as you said, it’s largely out of control.
English
@RosuGrigore @Mus_ab @chenxiaohong Recently, I've had a large number of evidently fake accounts with pi-squared in their name follow me here. A bunch of them are promoting FastSet. Do you have any thoughts as to where these come from?
English
@Mus_ab @RosuGrigore @chenxiaohong @pisquared The easiest way is for you to try to filter your following...
English
@PetarMax @RosuGrigore @chenxiaohong We’ve noticed that too. If you see fake accounts using our name or pretending to be from us or @pisquared, please flag them to us and report them to X so X can take them down. Thanks for highlighting this.
English
@RosuGrigore @pisquared and if I'm interested in removing what appears to be a fleet of pi-squared bot accounts that have started following me, whom should I contact?
English
Our amazing Matthieu! Head of BD. Contact him if interested in partnering or integrating with @pisquared.
Fast@PiSquared
Meet @Matt_Betton, Head of BD at Pi². He drives partnerships and strategy to connect the right people to what we are building. Based in Dubai, he loves travel, tennis, and live music. He believes Pi² goes further than blockchains with real speed, scale, and decentralization. That’s why he Pi’s.
English
Petar Maksimović | Nethermind retweetledi
1/ Today marks the beginning of a new chapter for Irreducible. We started as a custom-hardware company, and now we are fully focused on high-performance software.
Read below about our pivot and our new proof system, Binius64 👇
English