Daniel Rodriguez

Answered How do you handle a model that refuses a legitimate security question?

Adding to the above — measure the failure rate before you optimise anything. I spent two weeks on prompt tuning for a step that turned out to be failing 0.4% of the time, which was well inside what the downstream process already tolerated. Instrument first. The expensive problem is rarely the one that feels most annoying.

Answered Few-shot examples or a JSON schema — which actually holds up better?

Benchmarked on 150 tasks: temperature 0 and 0.3 were within noise on correctness. What did differ was diversity across runs — at 0 you get the same wrong answer five times, at 0.3 one of five attempts is often right. So it depends on your loop. Single-shot into production: 0. Generating candidates you filter with tests: higher, and take the one that passes.

Shared a bide Build an interview-ready STAR story from a messy memory

Turns a half-remembered work story into a STAR answer with 60- and 30-second spoken versions, likely follow-ups, and placeholders where facts are missing.

Shared a bide Refactor legacy code behind a characterization test gate

Make an AI or coding agent pin current behavior with tests, refactor in small steps, and never edit a test to make it pass, with preserved bugs listed.