Alex J Best (@AlexJBest) profile picture

Researcher in maths+formal methods+ml. Using formal verification to train models for hardware, software, mathematics, general reasoning @harmonicmath

London, UK
Joined May 2010
Alex J Best retweeted
I'm catching all these friends visiting my home while on the way back! Satnam's speaking in the CL right now!
A great talk by @danghica on how to civilize AI coding at Huawei. Some common approaches to the talk I will give later in the day. talks.cam.ac.uk/talk/index/2…
5
1
39
1,403
Alex J Best retweeted
I pointed Aristotle to egg, one of my favorite Rust projects, and asked it to prove a property over all egg runs on a simple arithmetic language. Pretty cool to see it formally reason about ~6kloc of a complicated e-graph library! aristotle.harmonic.fun/proje…
11
1
20
2,571
Alex J Best retweeted
We gave Aristotle the C code Bitcoin Core uses to parse signatures, and asked it to rewrite it in Rust and prove the Rust does what the C does 🧵 1/5
18
23
6
122
14,109
Alex J Best retweeted
🧵 Formal verification on @solana Absolutely thrilled by this 1st release. @HarmonicMath, founded by @vladtenev and @tachim, just added support for Rust to their product Aristotle, with Solana programs being a FIRST-CLASS citizen. They worked closely with 15 teams from ecosystem, calibrating Aristotle on the typical properties you'd expect from an on-chain program on Solana
29
46
19
308
107,287
Alex J Best retweeted
Rust support launched today! Before launch, we worked with @ManifestTrade on their open source order book on Solana. Their lead developer said the proofs "...go beyond the normal proof of no loss of funds and show un-brickable" Read the case study here: harmonic.fun/aristotle/case-…
7
9
3
51
4,721
Alex J Best retweeted
Our Aristotle AI theorem prover now works on Rust (including unsafe blocks), in addition to SystemVerilog (SVA or auto-formalized properties) and C (memory safety). Give it a try! harmonic.fun/aristotle/case-…
Aristotle now formally verifies Rust, joining C and Verilog 🦀 Prove your Rust code has no undefined behavior and does exactly what its spec says, even if you use unsafe. And if you’re porting C to Rust, prove the new code behaves the same as the original.
2
11
114
3,413
Alex J Best retweeted
Who would have thought?! Well, everyone, really - so bring your Rust code over, safe and unsafe alike!
Aristotle now formally verifies Rust, joining C and Verilog 🦀 Prove your Rust code has no undefined behavior and does exactly what its spec says, even if you use unsafe. And if you’re porting C to Rust, prove the new code behaves the same as the original.
3
3
25
749
Alex J Best retweeted
Aristotle now formally verifies Rust, joining C and Verilog 🦀 Prove your Rust code has no undefined behavior and does exactly what its spec says, even if you use unsafe. And if you’re porting C to Rust, prove the new code behaves the same as the original.
31
36
22
386
151,816
Alex J Best retweeted
formal verification is coming to solana @harmonicmath is bringing aristotle to rust they're building verified lean models of solana's runtime that will eventually be checked against the actual code great talk by @petarmax at scale or die
10
10
94
8,093
Alex J Best retweeted
bunnie put his security chip into the palm of my hand. I handed him back a @HarmonicMath Aristotle formal verification of his chip. You can read one of the verification reports here (proof of conformance with the SAIL formal specification of M-type RISC-V instructions): aristotle.harmonic.fun/proje…
9
21
2
191
8,629
Alex J Best retweeted
This Saturday, our formal verification engineer Petar Maksimovic (@petarmax) takes the stage at Scale or Die in London. See you there, @solana_devs!
4
12
3
44
4,272
Alex J Best retweeted
This thread contains links to Aristotle formal verification reports for a systolic array matrix multiplier at sizes 4x4, 8x8, 16x16, 32x32 and 64x64 for FP16 values. You can read the properties that were formally verified in Lean 4, see the SystemVerliog code, and read the verification summary. The formal verification is a direct comparison against a high level functional specification of a pipelined systolic matrix multiplier. I think this kind of direct formal verification of a matrix like this against a fairly direct high level specification is unusual.
9
18
112
4,993
Alex J Best retweeted
Your first $500 of Aristotle is on us. Pick the C function or Verilog module you trust least, and find out whether you were right to worry. aristotle.harmonic.fun
2
6
2
43
3,853
Aristotle can now formally verify Verilog and C! On @TensordyneInc's most complex block, with 12.2M gate equivalents, 1.24M flops and hundreds of state machines running concurrently, it proved 185 whole-block properties and found bugs up to 110,000 cycles deep.
2
9
44
1,577
Alex J Best retweeted
Tensordyne gave Aristotle its hardest block to verify: 12.2 million gate equivalents. 185 properties proved, 56 refuted. Jonathan Ko, their verification engineer, calls it "unthinkable with traditional tools." Read the the full @TensordyneInc case study: harmonic.fun/aristotle/case-…
3
19
2
99
9,091
Alex J Best retweeted
oh my. pointed aristitle at TorchLean's C runtime code and in ~6 hours: 211 of 213 properties proved (memory safety, refcounts, functional correctness) The other two came with a use-after-free counterexample involving inconsistent caller reference counts. We are reviewing the ownership assumptions and proof artifacts to establish exactly what’s guaranteed: ) im very impressed!
We’re bringing Aristotle to code. AI is writing more code than anyone can review. We're building the AI that verifies it. Starting today with C and Verilog.
2
10
2
39
3,635
Alex J Best retweeted
If you can verify math, you can verify code. And hardware. And pretty much any system you can model! "US ⁠fusion lab DIII-D said Aristotle rebuilt part of its reactor-control software in ⁠four ​hours from one research ​paper and a 254-word prompt." reuters.com/technology/robin…
We’re bringing Aristotle to code. AI is writing more code than anyone can review. We're building the AI that verifies it. Starting today with C and Verilog.
4
8
59
2,745
Alex J Best retweeted
Fast inference and verified answers go well together. Tudor (@tachim) is on the Neo Labs panel at the SGLang Summit with Igor Babuschkin, Eric Zelikman and Liam Fedus, Nov 12–13 in SF. Don't miss this one.
Proud to welcome our newest sponsors to SGLang Summit. In just one week since our public announcement, our sponsor lineup has grown from 11 to 36. Thank you to every sponsor helping bring this community together. Join 2,000+ engineers, researchers, and builders from across the AI ecosystem for technical sessions, hands-on workshops, and the GPU Expo. Hear from Lip-Bu Tan, Lilian Weng, Soumith Chintala, Aravind Srinivas, and more. Meet the teams building across AI hardware, infrastructure, models, and applications. 🗓️ November 12–13, 2026 🌉 Fort Mason, San Francisco 🎟️ Get your tickets: sglang.io/summit
10
6
1
37
10,462
Alex J Best retweeted
I mentioned Gödel and Escher during my talk at REBASE 2026 but I could not fit in Bach. conf.researchr.org/details/s…
9
9
2
137
8,066