# Zero_RL

- **Event:** [OpenAI Codex Hackathon](https://cerebralvalley.ai/e/openai-codex-hackathon)
- **When:** Thu, Feb 5 at 9:00 AM – 9:00 PM (PST)
- **Where:** Location TBA
- **Team:** [Sanjay Sai](https://cerebralvalley.ai/u/sanjay_sai), [Tarun Reddi](https://cerebralvalley.ai/u/TarunReddi)
- **GitHub:** https://github.com/REDDITARUN/zero_rl
- **Demo video:** https://www.loom.com/share/f2cd3eb1de894526b878b88f4e43eeba
- **Gallery:** https://cerebralvalley.ai/e/openai-codex-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/openai-codex-hackathon/hackathon/gallery/41

ZeroRL is an AI-powered RL environment generator built for the Codex Hackathon.
It solves the bottleneck of RL prototyping: users describe an environment in natural language, and the system generates a working Gymnasium environment, validates it, and prepares training/eval workflows with a live interactive UI.

---

Markdown version of https://cerebralvalley.ai/e/openai-codex-hackathon/hackathon/gallery/41. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt