Your mission is to make Kestrel and the AI systems it governs resilient against realistic adversaries, misuse, and control failure.
What you will own
- Model threats across product, cloud, identity, AI, and operational boundaries.
- Design and execute adversarial tests with reproducible evidence.
- Partner with engineering on prevention, detection, response, and hardening.
- Prioritise findings by credible impact rather than novelty alone.
- Maintain disciplined disclosure, remediation, and verification practices.
What we look for
- Hands-on security engineering, red teaming, cloud security, or adversarial AI experience.
- Ability to move from threat hypothesis to validated finding.
- Strong understanding of identity, access, data, and system boundaries.
- Clear evidence writing and responsible handling of sensitive findings.
- Sound judgment under uncertainty and incident pressure.
How we hire
No resume is required at this stage. We begin with structured questions so your judgment and actual work can carry the application. We do not use AI to rank, score, summarise, or reject candidates.