We pair machine learning engineers with medical students to tackle real clinical problems, guided by faculty physician mentors. Fellow applications for the Summer 2026 cohort are now closed, and selected teams are moving into the research cycle.
Many of medicine's biggest challenges are problems of data, prediction, and decision-making. Solving them requires both clinical intuition and technical fluency. But those two worlds often do not meet.
We created The Tensor Lab to address this. Our 10-week summer fellowship pairs talented machine learning engineers with medical students and practicing physicians to lead focused research. Fellows can work either independently or in small teams. The result is meaningful translational research.
This is a remote, project-based unpaid fellowship to train future innovators to design and lead medical research, from idea to execution. And they'll be in a cohort of equally driven, curious, and talented students to learn from. Successful projects will lead to academic publication and even influence how care is delivered to real patients.
You are the technical lead on the project, working alongside a medical student under the supervision of a faculty physician. Ten to fifteen hours per week from June through August, remote.
A medical student who framed the research question and helps you interpret results. Coordinates with the faculty physician and translates between the clinic and your technical side.
A practicing physician and published researcher who sponsors the project, secures data access, and sets the scientific standard. Frequent check-ins with the team.
Key dates for Summer 2026. The application window has closed, and selected teams are moving through onboarding and the summer research sprint.
The fellow application window ran from April 27 through May 26. We are no longer accepting new fellow applications for the Summer 2026 cohort.
Members of your team will get necessary credentials and access to the tools they need to get started.
Build the model, run experiments, analyze results, and write the paper. This is the core of the fellowship.
Teams present their final work and research findings to the rest of the cohort, faculty advisors, and the broader community.
Teams are supported in pursuing academic communication of results, including presentations at conferences and peer-reviewed publications.
No. Your medical lead and faculty physician advisor provide clinical context. The 2026 application cycle is now closed, but the program is designed for strong technical contributors to pick up the clinical side as they go.
Plan for ten to fifteen hours a week from June through August. Think of it like a serious research internship. If your circumstances shift, we can adjust.
You can expect one hour of curriculum plus a cohort meeting each week, on top of your project work with your team.
Remote by default. You meet your team over Zoom or Slack. Some chapters organize optional in person working sessions if there is geographic overlap.
No. The fellowship is unpaid.
We default to open science. Code is open sourced and all team members share authorship. Patient data stays under the physician advisor's institutional controls.
For the 2026 cycle, we considered current students, recent graduates, and career switchers with strong ML skills. The fellowship runs in the summer, but participation is not limited to people on academic calendars.
Applicants ranked three projects. If a first choice filled before an application was processed, the applicant stayed in the running for their second and third choices and could update preferences when needed.