Experience & Education.

Built a B2B customer churn prediction model on a 4 person team using an Explainable Boosting Machine (EBM) with expanding window temporal cross validation, optimizing for PR AUC and lift at the top decile while preserving feature level interpretability for stakeholder review. Engineered role segmented features across engagement, account tenure, and product usage signals; tuned classification thresholds against retention intervention costs. Delivered findings to Cox stakeholders to inform account management prioritization for at risk customers.

Fine tuned a local open source LLM on curated calculus problem sets to serve as an AI tutor, achieving over 95% accuracy on held out evaluation problems. Built a retrieval augmented generation system over FDA regulatory meeting documents, reaching roughly 80% retrieval accuracy through iterative chunking and embedding strategy refinement. Collected and processed FDA advisory committee transcripts for an ongoing study predicting medical product efficacy from regulatory deliberation quality; engineered NLP features to quantify discussion characteristics as inputs to downstream predictive modeling.

Managed $100K+ monthly procurement and supervised a team of nine. Improved order processing times by 50% through restructured vendor communication and tightened internal handoffs.

Lead end to end operations for Emory's flagship hackathon. Hosted 208 participants from 35 universities who built 40 projects in 2025. Manage a $10K+ budget and tiered sponsorship packages from $500 to $3,000 for the November 2026 event.

Coursework in Machine Learning (PySpark, MLlib), Causal Inference, Prescriptive Analytics, Big Data & Cloud Technologies, and Economics of Digital Platforms.

Bachelor of Business Administration with a concentration in Information Systems & Operations Management. Minor in Computer Informatics from Emory College.