
GitHub - Danau5tin/ai-trains-ai: RL-training an AI agent to RL-train AI agents.
RL-training an AI agent to RL-train AI agents. Contribute to Danau5tin/ai-trains-ai development by creating an account on GitHub.
Machine Learning Engineers and AI Developers
The repository provides ready-to-run training recipes for Fireworks AI covering reinforcement learning (GRPO, DAPO, GSPO, CISPO), preference optimization (DPO, ORPO), and supervised fine-tuning (SFT).
It includes fork-and-customize training loop scripts, shared utilities, worked examples, and a Fireworks Agent skill that handles data inspection, hyperparameter sweeps, training, and deployment from a single natural-language instruction. Standalone customer-facing scripts, integration tools, and a React-based renderer verifier round out the project structure.