reinforcement learning agents

Hands-on, bash-first playbook for AI gaming on Linux: set up a clean, reproducible stack, then build four weekend projects - RL with Stable-Baselines3 (CartPole and optional Atari), an MCTS Tic-Tac-Toe bot, an OpenCV+xdotool vision/keyboard bot, and a tiny ASCII dungeon generator - with exact commands, runnable code, and tips for logging, headless training, tmux, and publishing.