3 comments

[ 3.1 ms ] story [ 26.1 ms ] thread
Good write up. I wonder what approach an LLM would have taken for this optimization, say using autosearch.