Research Scientist, Formal Methods & AI Security

The role

You lead our research on AI agents that reason over confidential data across organisations without leaking it. You define what “no leakage” means, how to measure it and how to prove it. You also design the secure link between a private deployment on a customer’s own servers and our public cloud platform, which a distributed-systems engineer in the team builds.

Why this role

• An open problem where formal methods, cryptography and LLMs meet.
• Your theory goes straight into a working system, built by an engineer in the team.
• You lead the research agenda and write up the results for a peer-reviewed venue.

What you will do in your first year

• Define, formally, a measurable bound on what a debate between AI agents can leak, and build and validate an estimator for it.
• Design the link between the private and public sides: the protocol agents use to exchange arguments across it, and the proof that protected data cannot cross it directly.
• Design, with our CTO, how the checks that protect confidential data can be proven to an outside auditor.
• Set the measurement method, run the main leakage measurements and write up the results for a peer-reviewed venue.

What you bring

• A PhD (or a submitted thesis) in computer science, security, cryptography or a related field.
• Formal methods or information-flow control: non-interference, security type systems, protocol verification.
• Statistics for estimation: confidence bounds, variance reduction.
• Publications at strong security, formal-methods or machine-learning venues.
• Enough Python to build and run your own experiments.

Nice to have

Zero-knowledge proofs; multi-agent LLM systems; Tamarin, ProVerif, TLA+, Lean or Coq.