A curated list of autonomous improvement loops, research agents, and autoresearch-style systems inspired by Karpathy's autoresearch.

Build autonomous experiment loops that edit files, run tests, and keep only improvements for any project type

Pure RL Agents: I implemented Q-Learning agents that learn through Self-Play. They play against each other to get smarter without human help! Symmetry Optimization: To make them "genius" faster, I added logic so they understand that a board mirrored left-to-right is the same situation. This cuts the learning time in half!

Compare AdaL and Claude Code on Autoresearch benchmarks to find better hyperparameters, run more experiments, and converge faster