Training Agents: Live tutorial on how to fine-tune a coding agent for continual learning

Short summary

Hugging Face live tutorial on fine-tuning coding agents using supervised fine-tuning (SFT). Covers converting agent traces to training data, running TRL + LoRA fine-tunes on HF Jobs, and interpreting initial metrics. Includes reproducible repo and best practices for post-training workflows.

• •Learn SFT-first approach for agent training before advancing to GRPO or environment RL
• •Convert public agent traces into training examples and run TRL fine-tunes reproducibly
• •Understand what eval metrics can and cannot prove in early post-training phases

Generated with AI, which can make mistakes.