I'm a Senior Data Scientist at the University of Chicago's Biostatistics Laboratory, where I specialize in processing complex, unstructured data—from clinical text to time series analysis—with a focus on mental health and neurology research.
But here's the thing: I love music just as much as I love data science (okay, maybe music slightly more). After years of applying advanced statistical methods and machine learning to understand complex patterns in health data, I realized the same analytical rigor could revolutionize how we discover music.
Traditional recommendation algorithms trap you in bubbles based on your listening history or generic user profiles. Stell-R takes a completely different approach—I've built it to map the actual semantic and sonic relationships between artists, creating organic influence pathways that let you discover music based purely on content, not past behavior.
The same skills I use to analyze neural networks and estimate treatment effects in clinical trials now power the knowledge graph behind Stell-R's recommendations. It's research-grade data analysis applied to something I'm genuinely passionate about: helping people discover their next musical obsession.
My work focuses on unstructured data, end-to-end prediction pipelines, and causal inference—techniques that translate perfectly to understanding the complex web of musical influences and connections that traditional algorithms miss.
When I'm not building prediction pipelines for health research or refining Stell-R's algorithm, you'll find me volunteering at The True Vine Record Shop in Baltimore, always on the hunt for my next great musical discovery.
This hands-on experience with music curation informs Stell-R's philosophy: technology should enhance, not replace, the serendipitous joy of finding music that moves you.