
Fractional Member of Technical Staff, Autonomous Experimentation
Job Description
About the role
We're seeking exceptional Members of Technical Staff to push the frontier of what AI agents can do in the wet lab. AI has made it easy to generate hypotheses and experimental designs, but checking them against biological ground truth is still slow, manual lab work. You'll develop the methods, evaluations and training strategies that let agents design experiments, run them on lab instruments and interpret the results, dozens of experiments at a time.
Key responsibilities
- Design and run experiments that improve how agents turn a scientific goal into correct work on real instruments
- Build evaluations and benchmarks that expose where models fail on physical lab tasks, then turn those failures into training data, new methods or research directions
- Explore post-training, reinforcement learning against simulated and real instruments, and agent designs for long, multi-step procedures
- Study how models should handle ambiguity: when to ask the scientist, when to decide, and how to learn from each lab's corrections
- Turn messy behavior in real lab runs into concrete hypotheses, experiments and fixes
- Share what you learn through writing and publications where it helps the field
Minimum qualifications
- Have hands-on research experience with LLMs, such as post-training, RL, evaluations, agents or program synthesis
- Can move from a vague behavior problem to a concrete experiment: form the hypothesis, build the pipeline, run it, analyze the result and decide what's next
- Write strong Python and fluent in modern ML tooling
- Are excited by open-ended problems where the signal is noisy and progress takes both research taste and engineering execution
- Communicate clearly with scientists and engineers alike
Preferred qualifications
- Published research in ML, AI for science or program synthesis
- Graduate training in ML, computer science or a life science
- Experience with robotics, simulation or embodied agents
- Contributions to open-source ML or scientific software
- Experience as an early hire at a startup
The annual compensation range for this role is listed below.
Annual Salary (full-time equivalent): $180,000—$240,000 USD
Logistics
- Commitment: Part-time contract to start, with the option to grow into a full-time role.
- Location: San Francisco, CA preferred. Remote within the US considered.
- Equity: Meaningful.
We encourage you to apply even if you don't meet every qualification. We care more about what you've done than about checking boxes.
About the Company
Infera Biosystems is a Y Combinator-backed company building toward closed-loop science: AI agents that carry out research end-to-end in the wet lab, from designing experiments and running them on lab instruments, to analyzing the results and drafting follow-up work. Our mission is to close the loop between in silico work and the wet lab, so scientists can direct entire discovery programs and cure disease faster.
We're supported by some of the most prominent minds in biotech, and our founding team brings extensive experience across machine learning and bench science. One of our founders trained in biology and computer science; the other in mechanical engineering and computer science. Today we're a small team based in San Francisco, working with academic, hospital and pharma labs.
How we work
- We build with scientists, not just for them. We spend time in our users' labs, run their protocols with them and ship fixes based on what we see.
- We go after hard, real problems. Physical lab work is messy, and we'd rather solve it properly than demo around it.
- Everyone owns a lot. Each person on the team shapes the product, the customers we work with and how the company runs.
- We take safety seriously. Agents that touch physical labs need guardrails, and we look forward to working with diverse stakeholders to define alignment and safety with biological implications.
We're early, and the people who join now will shape what Infera becomes.