Want to know if this job is worth applying to?
San Francisco, California, United States
Remote only
Available
Optexity is a product-driven research lab building clinical reasoning and computer use models on data no other lab can reach. We deploy directly inside clinics and hospitals — including systems with no APIs, using integration infrastructure we've built and open-sourced — and in return become their preferred partner. That gives us proprietary clinical reasoning trajectories from practicing physicians, and a feedback loop between real patient encounters, our models, and the products built on top of them.
We're a small, fast-moving founding team with multiple published papers in NeurIPS, ICML, CVPR and backgrounds from Apple, Amazon, Microsoft, CMU, and IIT.
We are backed by world-class investors and leaders including Jeff Dean, Neotribe VC, PearVC, Together Fund, and Zapier Fund.
Customer obsession — we start with the customer and work backwards
Intellectual honesty — ideas matter more than titles; we communicate directly and assume good intent, even in disagreement
Bias for action — we build and learn with customers rather than debate in the abstract
Extreme ownership — we own outcomes, not just tasks, and see problems through
We're building a clinical AI assistant that sits inside real practices — surfacing care gaps, assembling pre-visit prep, flagging risk, and suggesting treatment plans. None of that works without physicians defining what "correct" means.
Most clinical advisor roles are a monthly call where someone reacts to a demo. This one isn't. You'll be in the product loop weekly: deciding what the assistant should flag, what a good summary actually looks like at 7:50am before a 15-minute visit, and grading where the model is wrong and why. Your judgment becomes our ground truth and our evaluation standard.
We're looking for physicians across internal medicine, cardiology, oncology, and other specialties — both to go deep in your own domain and to help us understand where workflows diverge across specialties.
Define what the assistant should surface. Which care gaps matter, which risk signals are worth an interruption, and which are noise that erodes trust. Translate clinical shorthand into precise, checkable criteria.
Shape the pre-visit summary. What belongs in it, in what order, at what length — and what a physician actually reads versus scrolls past.
Specify clinical search. What you'd want to pull from the chart before a visit, how you'd phrase it, and what a complete answer looks like.
Judge the AI. Review model-generated risk flags, differentials, and treatment plans against your own reasoning. Score them, explain the gap, and tell us whether a miss is dangerous or merely different.
Build ground truth. Help us construct real-case evaluation sets and rubrics that reflect actual standard of care, not textbook answers.
Pressure-test the product. Sit in on design decisions and tell us when something wouldn't survive contact with a real clinic day.
Contribute to publications. We publish benchmarks and technical reports; clinical co-authorship is available and encouraged.
MD or DO, board-certified or board-eligible
Currently practicing or recently practicing — active clinical context matters more than titles
Internal medicine, cardiology, oncology, or an adjacent specialty; primary care and multi-specialty experience especially welcome
Day-to-day fluency in an EMR (Epic, eClinicalWorks, Athena, ModMed, or similar) and an honest view of where it fails you
Opinionated about clinical workflow, and able to explain why something is right — not just that it is
Comfortable with ambiguity, direct feedback, and being asked "how do you actually know that?"
Genuinely interested in AI in medicine, including its failure modes
Clinical informatics training or CMIO-adjacent experience
Experience with quality measures and value-based care programs (HEDIS, MIPS, CPT Category II, HCC coding)
Prior work advising or building health tech products
Research or publication background
Experience training residents or fellows — the skill of articulating clinical reasoning out loud transfers directly
Part-time: 5–15 hours/week, flexible around clinical schedules. Async-friendly, with a recurring weekly working session.
Full-time: for physicians ready to step into a founding clinical role and own clinical direction end to end.
Remote is fine. Occasional in-person time in San Francisco or at partner clinic sites is a plus.
Direct, daily work with the founders — no layers between your input and what ships
Real influence over a product that practicing physicians will use, not a pilot that dies in committee
Competitive hourly rate for part-time engagements; competitive salary and meaningful equity for full-time
Co-authorship on benchmarks and technical reports
Visa sponsorship available for full-time roles
