amannnnn/baba-rlvr
0
Add hackathon blog writeup
fix docker file
push changes
pivot to pybaba, add docs, restructure
play: browser-based human-play mode at /play with arrow-key controls + BFS-solve button
viz: PIL-based trajectory visualizer + CLI (frame/play/solve/strip)
scaffold: engine, openenv server, reward tracker, MAP-Elites PCG, GRPO trainer, tests
