Step 2 of 9 · planned
Score one thing
Turn a hunch into a number. Write one scorer in plain code and run your first experiment.
What you'll build
A scorer that checks every number in a report appears in the day log, and an Eval() over the four dogs.
Braintrust features
What you'll be able to show
“67% — and here's the exact row that failed.”
🐾 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.