I'm a Data Analyst with 5+ years across academic medical centers — MedStar Health, Moffitt Cancer Center, Huntsman Cancer Institute. I work on clinical and EHR data: data analysis, predictive modeling, and shipping models into things people can actually use.
ISCVAM - Multi-Study Oncology Data Pipelines — Built two R pipelines and the Slurm orchestration behind a multi-study cancer analytics platform, automating ingestion, QC, and cell-type annotation across 198 studies (~6.3M cells).
Osteoporosis Risk Classifier — Balanced classes do not guarantee balanced predictions. On a perfectly balanced matched case-control study, every ML model family tested recalls controls more reliably than cases. The evidence so far points to heterogeneity among cases, not class imbalance or model choice.
Hiring Intelligence — Gradient boosting model for hiring outcome prediction, deployed live on Streamlit. Repo
VAERS Association Rule Mining — Apriori over 17,574 COVID-19 adverse event reports, surfacing symptom co-occurrence patterns.