Hi, I’m Jonathan Lai — a data scientist and ML engineer interested in building interpretable, human-centered systems. I’ve led projects from developing custom LLMs in PyTorch to studying political self-censorship on campuses, with experience across machine learning, NLP, statistical modeling, and full-stack development (Rails, React, Django).

I work end-to-end across the data science pipeline—from data collection and cleaning to modeling, deployment, and communicating insights through intuitive applications. Some of my recent work includes building patient-facing clinical dashboards with XGBoost + SHAP explainability, large-scale streaming pipelines at Vooks, and experimentation frameworks at VINCAR.

I first got into data through basketball analytics, and that curiosity still drives me today. I love uncovering structure in messy real-world systems and turning it into tools that help people make better decisions. Outside of work, I’m a family-focused person who values time with the people who matter most, something that keeps me grounded and reminds me why I build things to make life a little better for others.

Feel free to explore my projects or get in touch!