I'm a data scientist focused on real-world execution of AI. Currently a Research Scientist at TII, I develop methods to evaluate textual and agentic systems across search, conversation and tool use. My approach to evaluation is rooted in a decade of solving high-stakes data problems. From engineering aerial defense algorithms to deploying clinical decision support models that outperformed physicians, I know what it takes to ship reliable models into production.
Writing
See all posts-
Uplift Modeling IIPart 2: Decision making and evaluation metricsAug 2024 -
Uplift Modeling IPart 1: Motivation & Causal ModelingJun 2024 -
A Simple Hotel RecommenderBuilt with LLMs and TripAdvisor reviewsJun 2023 -
Predicting recipe popularityFor my favorite food blogOct 2022 -
Revisiting the bias variance tradeoffA classical principle in modern timesJul 2022 -
On the Hazards of Proxies IIPart 2: Prediction modelsApr 2022
Publications
- Comparison of Initial Artificial Intelligence (AI) and Final Physician Recommendations in AI-Assisted Virtual Urgent Care Visits Annals of Internal Medicine, 178(4), 498–506
- Using Deep Networks for Scientific Discovery in Physiological Signals Machine Learning for Healthcare (MLHC) 2020