I'm a medical student at Harvard and MIT. Previously, I studied Statistics and Biochemistry at Yale. I think a lot about performance and equity in health algorithms, especially for consumer wearables (Apple), laboratory medicine (PathAI), and clinical decision-making (HMS). My first-author research and perspectives have been published in the New England Journal of Medicine, JAMA, and more. Always happy to chat with fellow students interested in machine learning + medicine! Drop me a line at jdiao@hms.harvard.edu.


I grew up in Sugar Land, Texas, a suburban outgrowth of Houston. I spent most of high school with the speech and debate team, where I first started learning about public policy and ethics. I spent the rest of my time volunteering and working at the Texas Medical Center, before coming to the Northeast for college.


I graduated from Yale, where I studied Statistics and Biochemistry. My coursework inspired me to explore research in biomedical informatics, which led me to Gerstein Lab at Yale and Zak Lab at Harvard. Outside of class and lab, I spent most of my time learning ballroom dance, still my favorite hobby.

Medical School

Now at Harvard, I joined a subgroup of medical students dual-enrolled at MIT in the HST program. I recently finished my preclinical coursework with this amazing community of aspiring clinicians and researchers, and will be starting my rotations at Beth Israel Deaconess Medical Center in April 2021.


As a first-year medical student, I had the opportunity to join the tech startup PathAI and its vision to bring computer vision to pathologist workflows. As part of the machine learning team, I worked on deep learning for tissue and cell prediction, as well as interpretable prediction of molecular phenotypes.


For my research year, I joined Apple's Motion Health team, where I was the Directly Responsible Individual on several projects for the Apple Heart and Movement Study and LumiHealth, the Apple-Singapore national health initiative. My work focused on validating new health features across diverse populations.