Friday at 2 PM Pacific:
@dexhorthy and I are live-streaming a discussion about the (incredible) results from the new SlopCodeBench benchmark, and what it means for engineers.
Completely fascinating.
Register to attend here: gauntletai.com/ai-first-org/n…
- Nice to see SlopCodeBench the #2 topic on HN right now, right under Dario's statement. Go @GOrlanski! repo: github.com/SprocketLab/sl… paper: arxiv.org/abs/2603.24755
- I really need a language with the low-level control of Rust and the proof capabilities of Lean.
- It took me a few hours to write a Lean formalization of the theorems and proofs in a recent paper of ours. I used Fable + 5.6 Sol. I found the exercise shockingly eye opening. I don't think we should publish papers without Lean proofs at this point.
- in 1948, Turing introduced "unorganized machines", machines that learn, which we now call neural nets. in 1949, in "Checking a large routing", Turing introduced formal verification (Floyd-Hoare triples). Today, neural nets do formal verification. Pretty wild.



