Fine-Tuning & Evals

Someone on your team says "the model keeps getting the tone wrong, let's fine-tune it." Nine times out of ten, that's the wrong first move. You probably have a prompt problem, not a model problem, and you're about to spend two weeks and a dataset budget to fix

8 lessons, each with runnable code in the browser.

  1. Fine-Tune or Just Prompt?
  2. What an Eval Actually Is
  3. Scoring Fuzzy Answers
  4. LLM-as-Judge
  5. Build an Eval Suite
  6. Building a Training Dataset
  7. Overfitting and Generalization
  8. Tracking Evals Over Time

Compilearn home