Lab 15: Public Eval Methodology Teardown

Series: AI Eval Labs · Phase: Publish · Depends on: Lab 01, Lab 02, Lab 03

Core scope

• [ ] Deep-dive 1: grading rubrics
• [ ] Deep-dive 2: judge prompts and calibration
• [ ] Deep-dive 3: statistical framework

Stretch

• [ ] Series retrospective post

Links

• Code: series/ai-eval/15-methodology-teardown
• Demo: coming soon
• Write-up: coming soon on markrox.ai