Step 5 of 9 · planned
Judge what code can't
Some failures don't show up in a regex. Use a model as a judge, then check whether you agree with it.
What you'll build
An LLM-as-judge scorer for buried medication and invented warmth, validated against your own ratings.
Braintrust features
What you'll be able to show
“Code catches invented numbers. A judge catches buried medication.”
🐾 Not written yet. Each step is written as the course is actually worked through, so it describes real scores, real regressions and real false alarms rather than a script guessed in advance.