- Constructed Python web-scraping
- ETL procedures for Hawks front office NBA draft preparation process
- Applied classification models in Python, including single and local logistic regression and neural networks, to millions of rows of basketball play-by-play data to build a win probability model with accuracy above 0.75
- Developed win-probability-added metric to measure how much a player contributed to team's probability of winning a game, for use in team tools and reports for coaches, scouts, and front office, leading to greater