Skip to content
Heidelberg AI Curriculum
Heidelberg AI
Curriculum

1Pick your portal

MatrixListJourney

2Your references

AI newsDailyChangelogDailyFAQTipsHow-tos
CanvasesGlossaryTools indexApp store (opens in a new tab)LLM index (opens in a new tab)Videos
A living curriculum for practical AI work.
  1. Home
  2. Tips
  3. Set eval criteria before testing

Tips

Set eval criteria before testing

Write observable pass conditions before reviewing a model response.

Continue with the source

Read Regression sets and evaluation ops

Read Governance, evidence and handover

CurriculumAll tracks

Heidelberg AI Curriculum · 2026