# autoresearch

Ole Lehmann shares a method that applies Karpathy's 'autoresearch' concept to Claude skills — letting an agent test, tweak, and improve your prompts automatically. His landing page copy skill went from 56% to 92% pass rate with almost zero manual work.

Karpathy's Autoresearch isn't trying to be a general AI scientist. It's a ruthlessly simple experiment harness: the agent edits one file, runs for five minutes, checks one metric, keeps wins, discards losses. The lesson? The best autonomous systems aren't the freest — they're the most constrained.

Andrej Karpathy shares how his autoresearch agent autonomously tuned nanochat's training config over two days, found ~20 improvements to validation loss that transferred to a larger model, and pushed the Time to GPT-2 leaderboard from 2.02h to 1.80h — about 11% better.