Projects

DigitalOC

Full stack ML-powered NFL play prediction system

Over the course of the Fall '25 semester, my team and I developed DigitalOC from the ground up.

• Collected and processed 40,000+ in-game situations and stats from the past 4 NFL seasons using Pandas • Trained Scikit-Learn Random Forest models achieving 91.7% accuracy on run/pass play predictions • Built a React.js frontend with a clean, intuitive UI for real-time play predictions • Developed a Flask REST API backend to serve model predictions with low-latency responses • Gained hands-on experience with full-stack development, team collaboration, and version control (Git)

Python React.js Flask Scikit-Learn html/css

PharmaSched

Automated scheduling system for pharmacies

Over dinner, I asked my Mom (who works at a pharmacy), if she had any redundant, time consuming tasks. She mentioned that she spent several hours curating the monthly work schedule for each of the 5 departments. So I made PharmaSched... (she came up with the name)

• Built a lightweight scheduling solution that reduced monthly schedule creation time from 4+ hours to under 10 minutes • Developed a React frontend with Tailwind CSS for a clean, responsive UI across 5 department views • Implemented a TypeScript backend with type-safe data handling and validation • Utilized a local JSON database for easy data sharing and syncing between pharmacy workstations—no server required • Optimized for fast load times (<500ms) and minimal memory footprint (~50MB) • Supports scheduling for 20+ employees across multiple shifts with conflict detection

React.js Tailwind Vite HTML TypeScript