Fine-tune your own model for free: Claude Code writes the code, Kaggle T4s do the training

ojasvi_yadav · x · 2026-09-21

NandhaKishorM shared a near-zero-cost fine-tuning playbook: instead of writing training code yourself, use an agentic harness like Claude Code or Codex to modify the fine-tuning notebook in the laya repo (7.2k stars), then train on free Kaggle GPUs.

Original post →

More from coding & agent

coding & agent channel →