LLM-Based Cloud Seeding Dataset and Causal Analysis

First author of "Structured dataset of reported cloud seeding activities in the United States (2000–2025) using an LLM", published in Scientific Data. Built PDF extraction pipeline with OpenAI integration to process 1,000+ scanned NOAA cloud seeding reports into a structured dataset, helping address the gap in cloud seeding data. Conducted within-site difference-in-differences analysis to estimate the causal effect of cloud seeding on precipitation: Streamlit dashboard.

Alexa+ Smart Display Experiences

Improved device engagement and customer satisfaction with Alexa's smart display devices. Launched and measurably improved the quality of the new multimodal screen experiences for Alexa+.

Product Science, LLC

Founded Product Science Consulting to help early-stage startups get from insights through measurable impact. Partnered with the founder of Mayah Design to improve product strategy, product development, and setup analytics infrastructure end to end.

DesignMyExperiment.com

Experiment design service and curated libraries of real product A/B tests and experimentation methodology reviews.

HealthEvidence.org

Health evidence analysis website to help people distinguish strong health evidence (early introduction to peanuts helps prevents peanut allergies) from weak claims (playing tennis will add 10 years to your life).

M.S. Data Science, Columbia University

• Courses in: Design & Analysis of Online Experiments, Causal Inference, Behavioral Economics
• Data Science Institute Scholar
• Columbia Build Lab Intern
• Guest Lecture @ CUNY Graduate Center: Using LLMs for Science (PDF)
• Behavioral Economics research proposal: Strategic Route Choice in Navigation Apps: Testing for Level-K Thinking (PDF)

B.S. Computer Science, George Mason University

• Computer Science Teaching Assistant and Peer Mentor