Security Engineer (AI Red Team)
Core
Test deployed AI systems and agents under adversarial pressure to identify security failures and help build stronger defences.
Role type
Senior IC AI security engineer (red teaming)
Builds
Reproducible test cases, small testing tools, and regression tests for AI security
Domain
AI security / Cybersecurity
Deliverable
production ML models
Required skills
Offensive security, application security testing, adversarial testing of AI systems, web application security, API security, authentication and authorisation, AI prompt engineering, code automation for testing
Preferred skills
LLM application assessment, agent workflow testing, security evaluation suite development, cloud and infrastructure security assessment
Technologies
LLMs, agents, APIs, cloud platforms
Responsibilities
Design and execute assessments of AI applications and agents, test realistic threats like indirect prompt injection, build reproducible test cases and tools, analyse model behaviour and application design vulnerabilities, document findings and recommend mitigations, retest fixes with engineering teams
Seniority
Senior, hands-on IC
