I'm a computer science and public policy student at the University of Chicago. I'm passionate about solving complex problems related to social impact, and enjoy working across areas such as education, health, and international development. Before this, I worked on impact evaluations at Mathematica, where I collected, cleaned, and analyzed survey and administrative data for organizations like the Gates Foundation, U.S. Department of Labor, and Millennium Challenge Corporation. I then transitioned to Vera Solutions, where I designed and managed relational data systems for nonprofits and foundations, overseeing data migrations, building automation workflows, and creating reports and dashboards.

I currently work as a Research Assistant with the Data Science Team at the TMW Center for Early Learning + Public Health at the University of Chicago, where I support a validation study of the Luet—a wearable device developed by the center to capture conversational turn counts and other indicators of early childhood language development. I’m building a pipeline to parse, transcribe, and analyze Luet audio recordings using parallel processing. I’m also developing infrastructure to extract structured data from standardized academic and cognitive assessments, used to benchmark language development metrics in the study’s analyses. Lastly, I'm developing an internal web application to manage and process classroom videos, using AWS MediaConvert and S3 to handle video merging.

🎧 Your Spotify World – A web app built with Django and MySQL that lets Spotify users map where their favorite artists are from around the world. It pulls data from the Spotify, MusicBrainz, and Nominatim APIs. Check out the live version here — just message me first to be added as a beta user!

🎨 Chicago Parks District Activity Finder – A web app built with Flask and SQLite that addresses limitations with existing municipal websites, helping residents of Chicago easily find city-sponsored programming near them. Users can filter by distance, activity types, and age groups, using an interactive interface with Leaflet maps and geolocation. The app scrapes real-time data from the Chicago Park District API, deduplicates activities, and supports paginated results with dynamic filtering. Check out the live version here.

📜 OpenPeru - Core team member of a civic tech project promoting government transparency in Peru. My role focuses on building a pipeline to scrape and extract key information from Peruvian congressional documents, building a compreshensive database of bills and other legislative actions. More on this soon!

🏡 Affordable Housing and Green Space Equity in Chicago – Used HTTPX to scrape open-source map data and reviews from Google and Yelp on parks and fields in Chicago, developed an accessibility index to quantify affordable housing units' access to high-quality green spaces in Chicago, and presented findings in an interactive dashboard through Dash.

🎤 Classifying Far-Right Extremism in Social Media – Trained a non-linear Support Vector Machine using scikit-learn to predict the likelihood that an English-language social media post expresses far-right extremist content. Engineered features include vector-based semantic similarity, text subjectivity, text toxicity, and the presence of profane or insider extremist terms. Manually labeled 4,670 posts to create training dataset. Deployed the model via a Flask web app to accept new text input and return predictions in real time.

🇹🇷 Mapping the 2023 Turkey Earthquake – Used ArcGIS Pro to analyze the effects of the 2023 earthquake on infrastructure and demography in Turkey. Mapped extensive building and road damage in affected areas, highlighted the displacement of millions, and emphasized the need for resilient infrastructure and post-disaster investment to aid long-term recovery.

• Opportunities to contribute to small/mid-sized organizations that focus on social impact.
• Mentorship and collaboration, especially in fast-paced, technical environments.

Let's connect!