Peloton Customer Churn Analysis
Investigating why Peloton customers reduce usage — combining survey data with geocoding and U.S. Census demographics to uncover patterns in churn across age, income, and geography.
USDA Food Access Research Atlas
Analyzing food desert conditions in San Diego's North County by comparing USDA census tract data from 2015 and 2019 — tracking shifts in poverty, demographics, and SNAP benefits.
About These Projects
These projects were completed as part of graduate coursework at the University of Southern California, Viterbi School of Engineering. Each project demonstrates end-to-end data science workflows — from data collection and cleaning through analysis and visualization — applied to real-world questions about consumer behavior and public health.
All analysis was performed in Python using Jupyter Notebooks (Google Colab). Click into each project to explore the methodology, data pipeline, visualizations, and key findings.