Training a PPO Agent with Grok to Play a Self-Built Game
tetsuoai · x · 2026-09-01
The author used Grok to build a browser game called Roofline and then trained a PPO agent using Grok to play the live page. The best recorded run achieved a score of 39,359, covering 4,924 meters with a combo of 117. The game is available to play online.
More from coding & agent
- Mech engineer pivots to AI automation freelancing, seeks reality check on market saturation — noyescape · 2026-09-01
- Obsidian-skills launches, turning your note vault into an AI agent playground — bigaiguy · 2026-09-01
- hermes-desktop Removes Terminal Barrier for Hermes Agent — thisguyknowsai · 2026-09-01
- Securing AI coding agents: preventing injection and leaks in internal codebases — West_umxwareness4908 · 2026-09-01
- Automate Academic Research Pipeline with Claude Code — Imbad0202 · 2026-09-01
- OpenClaude: Universal AI Agent CLI Running Anywhere — Gitlawb · 2026-09-01