Article URL: https://github.com/Danau5tin/ai-trains-ai Comments URL: https://news.ycombinator.com/item?id=48905919 Points: 11 # Comments: 0

🔓 Everything is open sourced including: the trained agent's weights (LoRA adapter on 🤗 HF), agent harness, task families, reward code, GPU orchestration, tinker RL training scripts, and retro write-ups of every pilot (including the failures). Jump to Getting started ↓ An AI in an RL loop, whose action is training AI in an RL loop. (Source: assets/hero.svg.) Tinker trains the agent. The agent writes verifiers envs, rubrics, and prime-rl configs. prime-rl trains the small model. The inner model's hidden-eval score flows back up as the outer loop's reward. One episode = one attempt by the trainer agent to produce a valid, high-quality training job for a given task: The outer loop then RL-trains the agent itself on episode reward, using Tinker. Every outer-loop batch spawns 40 real inner training jobs across up to 16 GPU pods. Note for close readers: the agent-facing prompt (template/INSTRUCTIONS.md) gives the agent a simplified view — the 75/25 uplift/absolute split inside job quality, plus a fewer-attempts nudge — not the full 0.35/0.60/0.05 decomposition. The published adapter was trained against that prompt; the reward actually computed is the one above. Six families of tasks, deliberately built so that the untrained models struggled without training, and all require multi-step tool use and reasoning: Five families train the agent; triage is never trained on and serves as the generalisation probe.