Building Mathematical Superintelligence
Joined January 2024
- Tweets328
- Following8
- Followers27.5K
- Likes325
Fifty years of C and we're still shipping buffer overflows.
Aristotle proves your C does what you said it should. When it doesn't, you get the input that breaks it.
New sign-ups get $500 in credits. Get started with the most effective way to protect your enterprise:
aristotle.harmonic.fun/?r=x
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-…
C and Verilog are just the start. More languages are on the way.
Try Aristotle today: aristotle.harmonic.fun/
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
Harmonic retweeted
@HarmonicMath is hiring a Formal Verification Engineer to develop Aristotle and apply it to formally verify production hardware and software!
The work spans formal methods research, product development and customer deployments.
Join our team in Palo Alto or London!
🔗 jobs.ashbyhq.com/Harmonic/5e…
Harmonic retweeted
Today we're launching Miles v0.1, an open-source RL framework for LLMs and multimodal models.
RL training is easy to start and hard to debug. Miles helps you ensure your run is correct, use hardware efficiently, and keep RL running at scale.
Over the past 9 months, 72 contributors have landed 1,326 commits, 85 GPU E2E CI tests, battle-testing Miles on frontier open models like Kimi K3, DeepSeek V4, Qwen 3.8, GLM 5.2, Inkling, MiniMax H3, etc.
Miles powers frontier-model development and production RL workloads at @humansand, @periodiclabs, @modal, @DecagonAI, @Eigent_AI, @nebiusai, @IBM and more, on both @NVIDIAAI and @AIatAMD hardware.
Here is what we built, and why teams picked Miles🧵
Aristotle
I am absolutely stunned by how effective @HarmonicMath is at lean 4 formalization of mathematics. It's able to fill in gaps and understand intent as well as intelligently leave out of scope work axiomatized when automatically formalising math.
🔥Sir Timothy Gowers shares how he used Aristotle to effortlessly autoformalize a complex paper into Lean—without needing to know a single line of Lean himself.
The future of mathematical research is changing fast.
Read more:
gowers.wordpress.com/2026/07…
Harmonic retweeted
Luckily, I had recently been contacted by @PietroMonticone from harmonic.fun, who asked whether I had any formalization projects I would be interested in undertaking using their Aristotle system. I'm happy to say that it has just successfully formalized our paper.
Harmonic retweeted
A pleasure to help @wtgowers and his colleagues get their preprint formalised with Aristotle (@HarmonicMath).
The long case analysis that a referee would understandably prefer not to check by hand is now formally verified in @LeanProver.
The code should be available on @GitHub before long. After some further cleanup, parts of the supporting API may also be suitable for upstreaming to Mathlib.
AI is accelerating mathematical research in increasingly deep problems
We’re partnering with the @AIMathematics to develop open, human-centric benchmarks for AI in mathematical research
The benchmark is built around 50+ hard problems selected by mathematicians to measure how AI can accelerate discovery, from generating new ideas to tackling some of the field’s most challenging questions
From @emilyriehl on the #aboutlogic podcast (w/ @DenizPhiMa):
"I much prefer to interact with an autoformalisation agent than a large language model in discussing mathematics, because the large language models will feed you a lot of bullshit ..."
"I came up with my own counterexample, and then I asked Harmonic's agent Aristotle to verify it for me in Lean as a kind of extra check that my counterexample was correct. I did not ask a large language model for this, because if they tell me it's correct, that gives me no assurance ..."
"Aristotle was able to confirm that it is a valid counterexample."
We are very happy that Emily Riehl @emilyriehl joined or little podcast #aboutlogic and talked about higher categories, synthetic mathematics, formalization and much more.
youtu.be/4MQbd5wTlI8?si=0QL_…
🪄🧙Aristotle is the mathematician's super-assistant
Check out how @LorenzoLuccioli uses Aristotle to develop new results in algebraic combinatorics
Happy to share that our paper
“Mapping Uncharted Symmetries: Machine Discovery in Combinatorics”
has been accepted to the ICML 2026 AI4Math Workshop.
We study AI for discovery in algebraic combinatorics, with verification in @leanprover using @HarmonicMath’s Aristotle.
1/11