Projects
Here are selected projects from my career in data science and analytics:
LSTM-Based Anomaly Detection for Checkout System Monitoring
Developed a deep learning anomaly detection system to proactively identify checkout system degradation before guest-facing outages occur, combining LSTM autoencoders with statistical process control.
View project →Agent-Based Marketing Mix Simulator for COVID-era Planning
Designed and implemented an agent-based simulation model to preserve marketing mix optimization capability when COVID-19 rendered traditional regression models unusable due to structural breaks in historical data.
View project →Live Cell Tracking in Oscillating Intensity Images
Designed and implemented a novel computer vision system for tracking biological cells in microscopy images with time-varying intensity, using particle filters, optical flow, and finite elements simulation. 25,000 lines of cross-platform C++.
View project →Propensity-to-Purchase Modeling Framework
Architected and validated a multi-brand propensity modeling system enabling personalized marketing experiments across Disney's parks and experiences divisions, achieving 0.85-0.90 AUC through rigorous model validation and quality assurance.
View project →Automated Traffic Forecasting for Capacity Planning
Built an automated multi-endpoint forecasting system enabling infrastructure teams to proactively scale resources ahead of high-traffic ad campaigns, achieving ~5-10% MAPE across 90-day horizons.
View project →