Last update from the experiment - after more careful prodding, it contains more real implementations of the prover components, including a claimed bidirectional fft, with updated benchmarks, which are slower Most of my involvement has been more of the sort of “is it a real implementation? Are you sure? Don’t take shortcuts and work for a day if needed” Which brings the question of how well would an alternative “middle management for LLMs” (as @alexhevans called it) work here 😅
Kobi Gurkan
Kobi Gurkan5.7. klo 09.56
ok a bit embarrassing - I didn't have rust installed on that machine, so the generated code didn't even compile and the benchmarks were estimates, as @MatteoFrig60839 also noticed I prodded more to get it to a better state, but still not a fully working implementation I would still be pretty suspicious if it's both correct + faster than a carefully hand crafted production grade C++ implementation the ultimate test of having a rust generated proof verified in a c++ verifier - I can't get right for now and would require more prodding would be nice if we had a good isolated full proof verification function in C++ and point CC towards that
1,82K