Work

I am exploring consequential problems across healthcare and life sciences, especially where fragmented workflows and data create opportunities for better software and applied AI.

Built model training, release, personalization, and failure-analysis systems for Meta's neural-input wristband, including release automation that cut roughly two weeks of intensive engineering work to a few days of light supervision.

Led a 14-person engineering and product team building patient matching, record reconciliation, terminology normalization, and search infrastructure that unified millions of records from thousands of healthcare sources.

Built model development and deployment infrastructure for more than ten data scientists, enabling models to reach production in weeks, and shipped systems supporting coaching insights, data quality, and measurement of participant outcomes.

I completed my Master's degree in Computer Science with a concentration in Data Science at UMass Amherst, while working full time at Omada.

Built APIs for training and serving computer-vision and language models in Salesforce Einstein, and co-invented a patented system for generating augmented neural-network training datasets.

Built reactor lifecycle simulation software used to evaluate next-generation nuclear reactor designs.