# Research Scientist, Medical World Models at Function Health

- Company: Function Health
- What the company does: It’s time you own your health. Function includes 160+ lab tests and personalized protocols for instant action. Tracked over time in one secure place. Backed by Battery, General Catalyst and a16z.
- Company website: https://www.functionhealth.com/?utm_source=google&utm_medium=Google&utm_campaign=20800990020&utm_content={adid}&utm_term=function%20health&gad_source=1&gad_campaignid=20800990020&gbraid=0AAAAAqiw_KFExiKCoQ9G5NEKwcrrjePXb&gclid=Cj0KCQiAi9rJBhCYARIsALyPDttmpuDOephC5Um1LdEhfu26AqPmxZ67mRocG5lC6iFX7x2YcWSfeKMaArUgEALw_wcB
- Type: Startups (AI role)
- Level: Mid level
- Location: United States - Remote
- Work setup: Remote
- Posted: 2026-09-21
- Apply by: 2026-11-05
- Apply: https://jobs.gem.com/function-health/am9icG9zdDp51wm2xDbpq2IlLL0WGfgH
- Page: https://www.1752.vc/careers/jobs/function-health-research-scientist-medical-world-models/

## About the role

You are a researcher who likes to ship and/or an engineer who insists on evidence. You have trained models on messy, irregular, real-world data and know that the evaluation design usually matters more than the architecture. You are comfortable being early; defining the problem, the dataset request and the metric before the first training run, and you communicate clearly with clinicians and regulators as well as with ML peers. You care that the model behaves well for the person on the other end of it.

## What they're looking for

- PhD in machine learning, computer science, biomedical engineering or a related field with 1-2+ years of professional experience, or MS/BS with 5+ years building and evaluating ML models on real data
- Strong Python and PyTorch, experience training at scale (multi-GPU, large datasets, experiment tracking) and writing code others build on
- Rigorous evaluation instincts: statistical thinking, calibration, error analysis, and the habit of asking whether a result would survive a larger validation set
- Publication record at top ML or medical-imaging venues (e.g., MICCAI, NeurIPS, ICML, ICLR, CVPR), or equivalent evidence of research output delivered into production
- Clear written and verbal communication with technical and clinical audiences