Senior Research Scientist, AI Bio Capabilities Verification
- Project/Division: AI, Foundational Research
- Start date: As soon as possible
- Status: Full-time
- Work location: In-person, Bay Area
- Approximate compensation: $195,000 – $300,000, based on experience
- Reports to: Wet Lab Lead
The role
SecureBio is hiring a Senior Research Scientist to design and run the _in vitro_ capability-verification experiments that our Foundational Research team will use to establish how _in silico_ model performance holds up in real biology.
Our _in silico_ evaluations have become de facto standards, and frontier labs run them before releasing models to the public. Our work has been cited in model cards from OpenAI, Anthropic, and Google, and has informed safeguards on frontier models. We are a member of the NIST US AI Safety Consortium and a partner to the European AI Office.
It remains an open question whether such digital proxies can truly serve to underpin pivotal, consequential decisions regarding biological risks in AI. That is the question that you will seek to answer.
The team
Our Foundational Research team builds the measurement tools for capabilities that do not yet have established methods. The capability verification workstream complements our work in uplift and hybrid evaluation development.
This kind of work is new: AI just may accelerate biotechnology like semiconductors did computing half a century ago. We think this demands a research facility with the creativity, flexibility, and resources to keep pace; a sort of Bell Labs for biosecurity. We are looking for someone with strong research taste who can work with a diverse team to design rigorous studies from scratch and rapidly produce results that are clear and convincing to AI firms, researchers, and policymakers.
We are currently building a BSL-2 lab and expanding the team that will work from it. You would join early enough to have a significant opportunity to shape both of these.
Responsibilities
- Design and run _in vitro_ validations of specific capabilities that LLMs and biological AI models demonstrate computationally, on turnarounds short enough to inform pre-release decisions.
- Own experimental design rigor across verification studies: power, controls, bias mitigation, replicability, and confound management.
- Translate candidate verification tasks into executable experiments, including choosin...
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Sed ut perspiciatis unde omnis iste natus error sit voluptatem accusantium doloremque laudantium, totam rem aperiam, eaque ipsa quae ab illo inventore veritatis.