Project : SJSU RIOT

• Modernized SJSU RIOT classroom observation platform from legacy Python 2.7 to production-ready Python 3.11/Flask application deployed on Google App Engine, now serving 500+ daily concurrent users

• Architected secure multi-user infrastructure implementing OAuth 2.0 with PKCE authentication, namespace-isolated Google Cloud Datastore for data segregation, and auto-scaling configuration (0-10 instances) with 8-hour session management

• Developed real-time interaction tracking system with 17+ predefined observation categories, interactive timeline visualizations using Bootstrap/jQuery, CSV export functionality, and RESTful API endpoints for session recording and data management

Focused on large-scale spatial analysis and classification of ecological data to study fire patterns. Utilizing data science models, machine learning techniques, and geospatial tools to analyze and visualize fire behavior. Managing diverse datasets with Python, R, and GIS