Security engineer in AI infrastructure in London

We exist to make humanity more free. For most of human history, you farmed or you starved. Technology gave people more time for the things they wanted to do, instead of things they had to do. Powerful AI will be the biggest lever for human choice we've ever built - but only if models are aligned with what humanity actually wants. There are groups building AI who don't share these goals. Whoever deploys frontier compute infrastructure fastest will decide whether AI expands human freedom or shrinks it.

Responsibilities:

• Run red and purple team operations emulating advanced threat actors across our networks, data centre OT, and applications, extending the offensive program alongside the team's other specialists.
• Pen test the networks behind our GPU clusters, from frontend and backend fabrics to out-of-band management and site connectivity, with a focus on segmentation and tenant isolation.
• Test data centre OT, including building management, power monitoring, cooling controls, and the IT/OT boundary, in partnership with site operations so testing never puts a live site at risk.
• Break the customer-facing platform, including the console, public APIs, and identity and access controls, hunting for any path from one tenant to another.
• Assess internal and employee-built applications and the CI/CD pipelines that ship them, and turn recurring findings into secure defaults engineers adopt.
• Partner with security and engineering teams on AI-specific attack scenarios, including agentic systems and other novel AI-enabled risks, and contribute to AI-assisted offensive testing tools and workflows.
• Present findings to technical and executive audiences as clear, actionable risk that shapes the security roadmap.

Requirements:

• You have years of hands-on experience in red teaming and offensive security operations.
• You have deep expertise in at least two of: enterprise and data centre networks, OT/ICS, web and API applications, cloud (AWS, GCP, Azure), Kubernetes, CI/CD pipelines.
• You've found novel attack paths and chained vulnerabilities into full compromise, crossing network, application, and identity boundaries to get there.
• You've run red team operations and adversary simulations against well-defended environments.
• You've found authentication, authorization, and multi-tenant flaws in web applications and APIs through hands-on testing.
• You write quality code in Python, Go, or similar languages, and you build your own tooling when off-the-shelf tools fall short.
• You've put AI to work in real offensive engagements, using LLMs and agents to speed up recon, vulnerability discovery, exploit development, or attack-path analysis.
• You write findings that get fixed, and you've worked closely with blue teams to turn your operations into detections and hardening.

Preferred:

• State-nexus threat models (frontier AI, government, critical infrastructure).
• Data centre facility systems (BMS, EPMS, cooling controls, Modbus, BACnet).
• High-performance and data centre networking (InfiniBand, RoCE, BGP, out-of-band management).
• Testing AI and ML systems (agentic workflows, model serving).
• Defensive experience (detection engineering, SIEM, EDR).

Compensation Range: $224K - $275K