Coming soon
What good LLM evaluation actually looks like
Lessons from building evaluation pipelines in a regulated healthcare setting — what vibes-based testing misses and why it matters.
AI EvaluationLLM
§5 — Writing
Notes & Essays
On AI evaluation, causal inference, and doing rigorous data science in settings where it actually matters.
Coming soon
Causal inference for people who've done psychology experiments
A bridge for researchers moving into data science: what transfers, what doesn't, and where the real differences lie.
Causal InferenceMethods
Coming soon
The asymmetric cost of false negatives in safeguarding ML
Why standard accuracy metrics are the wrong frame for high-stakes classification, and how to think about it instead.
MLSafeguardingEvaluation