Offensive Security Engineer
UK AI Security Institute · London, UK · London
Research instituteonsite
At a glance
Highlights
- Attack evaluation sandboxes that contain frontier AI models during testing.
- Build agent-assisted red-teaming harnesses for continuous infrastructure testing.
- Publish research and talks on sandbox security and AI-assisted offense.
Ideal candidate
Experienced offensive security engineer who loves to build and break, eager to test the systems behind frontier AI evaluations and see findings shape how they are built.
Requirements
- Find and exploit vulnerabilities across cloud, infrastructure and application layers
- visa sponsorship
- Open to applicants of any nationality or background
- hiring process
- Apply by 15 November 2026; criminal record check and BPSS clearance required, CTC eligibility preferred.
Lab profile
About UK AI Security Institute
UK government institute studying advanced AI risks and turning research into protective action.
Known for
- Alignment red-teaming
- Control red-teaming
- Misuse red-teaming
- Cyber and autonomous systems security
Key facts
- Based in the UK
- UK government-backed
- World's largest dedicated advanced AI risk team
Posted · Last seen