Search Applied Research + Computer-Use Agent Research.

Prev trained Atlas Agent model, contributor to 5.2 + 5.4 agent capabilities.

Accelerated research by 40% for the National Institutes of Health (NIH) by launching data integration platform. Reduced LLM error rate by 10% via repairing AI clinical document analysis platform across 7500+ data tables. Led team of 5 engineers to increase internal platform usage by 15% by deploying LLM prompt iteration feature

Investigating efficient methods of prompting large language models (LLMs) such as ChatGPT. Mentored by Jesse Mu and Noah Goodman.