Skip to content
AI Engineer Guide.

Track

Evaluation & quality

Golden sets, judges, regression gates, and error analysis.

  1. Why evaluationconcept · T1
  2. Datasets & golden setsconcept · T1
  3. Coverageconcept · T1
  4. Rule metricsconcept · T1
  5. Semantic metricsconcept · T2
  6. LLM-as-judgeconcept · T1
  7. Human evaluationconcept · T1
  8. Inter-annotator agreementconcept · T2
  9. Rubricsconcept · T1
  10. Judge biasconcept · T2
  11. Judge calibrationconcept · T2
  12. Offline vs online evalconcept · T2
  13. Faithfulnessconcept · T1
  14. Answer qualityconcept · T1
  15. Context metricsconcept · T1
  16. Ranking metricsconcept · T1
  17. RAG metricsconcept · T2
  18. Task successconcept · T1
  19. Trajectory evaluationconcept · T1
  20. Tool-call accuracyconcept · T1
  21. Safety evaluationconcept · T1
  22. Adversarial evaluationconcept · T1
  23. Red teamingconcept · T1
  24. Online evalconcept · T2
  25. Regressionconcept · T1
  26. Benchmarksconcept · T2
  27. Eval-driven developmentconcept · T1
  28. Error analysisconcept · T1