Data Science. Artificial Intelligence.
My professional and academic experiences converge around a common interest: understanding complex systems, measuring what matters and using data to drive better decisions.
My foundation in Chemistry developed my appreciation for quantitative analysis, experimentation and evidence-based reasoning. Over time, that foundation expanded into a broader focus on performance evaluation, root cause analysis and continuous improvement. My transition to Data Science and Artificial Intelligence was a natural progression, applying the same analytical principles to increasingly complex systems and larger-scale problems.
What draws me to AI is not simply the technology itself, but the challenge of building systems that perform reliably and create value in real-world environments. That requires more than developing strong models. It requires defining what success means, understanding tradeoffs, assessing performance and using evidence to continually improve outcomes.
My graduate work in Data Science has given me the technical foundation to develop machine learning and AI-driven solutions, while my broader background shapes how I approach them. I think of AI not simply as models to optimize, but as systems to understand, evaluate and improve. This multidisciplinary perspective allows me to bring together analytical rigor, systems thinking and modern AI capabilities to solve complex problems.
A real-time gaze-to-speech tool that gives people with physical disabilities a voice using only their eye movements.
Using XGBoost to predict wildfire occurrence across 3, 7 and 14-day horizons.
Helping engineering and marketing teams get fast, audience-tailored answers from GenAI research documents.
A bidding agent for a simulated second-price ad auction that infers each user's value from observed behavior.
A research proposal examining racial disparities in clinician empathy during childbirth.
A multiple regression analysis investigating whether age predicts sleep duration, using CDC survey data.
A program that finds every valid word in a Scrabble rack, including wildcards, then scores and ranks them.
A convolutional neural network that reads retinal images and flags which patients should be referred for treatment.
A Neo4j graph of global refugee flows that uses network centrality to surface the top sending and receiving countries.
Interactive D3.js visualization investigating how caffeine intake relates to stress across 20 countries.