Data Scientist

Built end-to-end sports analytics pipelines and prediction models across NFL, NBA, and MLB, powering DFS projections, betting edge identification, and interactive performance dashboards.

Key Achievements

• •Engineered end-to-end NBA and NFL analytics pipelines using Python, Pandas, and AWS Lambda/ECS — aggregated player and team data from external APIs and proprietary datasets for real-time analysis
• •Developed NBA player projection models incorporating pace, defense vs. position, and cluster-based team tendencies to generate context-aware forecasts
• •Built DFS simulation and projection scripts automating daily updates with injury and rotation adjustments for DraftKings player pools
• •Created NFL fantasy scoring and betting edge models integrating bookmaker odds, coverage metrics, and custom play-by-play features to identify high-value prop opportunities
• •Developed a sophisticated NFL play-by-play data pipeline to extract nuanced insights like frequency and yards per route for receivers under varying coverage scenarios
• •Automated MLB data ingestion from FanGraphs, normalizing park and handedness factors for matchup analysis
• •Designed interactive dashboards (Streamlit, Tableau) for player and team trend visualization across all three sports
• •Leveraged Claude as an analytical co-pilot for NFL and NBA player analysis — using LLM-driven breakdowns to surface patterns and validate model outputs