I Spent 6 Weeks using Reinforcement Learning to Beat This Godot Game

Nick walks you through hooking up a Stable Baselines 3 agent to GoDotRL, diving into reward-shaping best practices and crafting rock-solid RL-friendly environments in Godot. All code, logs and 14 final game versions live on GitHub—just beware the 167 MB DMG files if you want the full build!

Along the way he’s sharing fun war stories (still trash-talking Claude), giving shoutouts to Ed Beeching and Brackeys, and plotting RL use in everything from mining to hard-mode platformers. If you’re in Florida, hit him up—otherwise, connect via LinkedIn, GitHub, Discord or Patreon and keep the RL hype alive!

Watch on YouTube