AI
Member of Technical Staff - Translational Medicine
Latch builds rigorous scientific benchmarks (benchmarks.bio http://benchmarks.bio) that measure how well AI agents perform real biology. Frontier AI labs and pharma use them to answer one question: can an agent handle the messy, multi-layer reasoning that real biological work demands?
Every benchmark is grounded in real experimental and clinical work, including designing experiments, interpreting data, reading clinical endpoints, and advancing compounds. Each is built by scientists who've done that work themselves.
We're roughly 95 people in San Francisco, backed by Lux Capital, Coatue, Fifty Years, Haystack, Jeff Dean, Alec Nielsen, and George Church.
We're seeking a Member of Technical Staff for Proteomics to lead our Proteomics benchmark and define how the field measures AI in this domain.
About the Role
You will lead the clinical side of what Latch measures. You will define the translational endpoints in our therapeutics and disease benchmarks and determine whether a model’s answer is correct in the sense that matters to a treating physician, investigator, or regulator.
You will also help obtain clinical data required to build these benchmarks. The most useful data is often not public. It sits within archived clinical trial records, patient registries, health systems, nonprofits, academic centers, and biotechnology companies. Access depends on trust, sound agreements, and a clear understanding of how sensitive data should be handled.
What You’ll Do
- Define and own the translational endpoints used across o