Free 🧰 Practical LLM Lessons

LLM Lab 8: Evaluate the Output

Score LLM responses for accuracy, completeness, and style using lightweight checks.

Published January 8, 2026

Previous

LLM Lab 7: Context and Grounding

Next

LLM Lab 9: Debugging Prompts

Set the rubric

You cannot improve what you do not measure. Define what a good answer looks like before you prompt.

Simple evaluation rubric

Correctness
Is each claim supported by the source?
Completeness
Did it answer every part of the question?
Format
Does it match the required structure and length?
Tone
Is the style appropriate for the audience?

Can You Guess?

You ask for a 3-bullet summary with citations. The model returns 5 bullets and no citations. How is the quality?

Try this now

Add a grading pass

  • Prompt for a summary with specific constraints (length, citations).
  • After the reply, ask: “List any ways you missed the constraints.”
  • Then ask it to deliver a corrected version based on that self-evaluation.

    • Key Takeaways

    • ✓ Define quality criteria up front.
    • ✓ Check correctness, completeness, format, and tone.
    • ✓ Use the model to self-evaluate before you accept an answer.

    Quick poll

    What brought you to this free lesson?

    Lesson: LLM Lab 8: Evaluate the Output

    ← Previous Lesson LLM Lab 7: Context and Grounding Next Lesson → LLM Lab 9: Debugging Prompts
    Back to Library Unlock More Lessons