Vibe Coding Forem

Vibe YouTube
Vibe YouTube

Posted on

Nicholas Renotte: I Spent 6 Weeks using Reinforcement Learning to Beat This GoDot Game

TL;DR

Nick spent six weeks wiring up a Stable Baselines 3 agent to GoDotRL, sharing his hard-earned tips on reward shaping and crafting RL-friendly GoDot environments. All code, logs and 14 iterations of the game are on GitHub—just ping him if you need the hefty DMG files via Git LFS.

He got the idea after seeing RL used for mining optimization at a think tank, so he built a fiendish game to test it. Big shout-outs to GoDotRL creator Ed Beeching and Brackeys, plus links to his socials if you want to follow along. Happy coding!

Watch on YouTube

Top comments (0)