AI Security Engineer
Core
Design, implement, and operate offensive cyber-capability evaluations for frontier, preview, production, and open-weight AI models, including vulnerability analysis, exploit development, and attack-path reasoning.
Role type
Senior IC AI Security Engineer (Offensive Evaluation)
Builds
Cyber ranges, vulnerable applications, exploit-development targets, and evaluation harnesses for autonomous AI systems.
Domain
AI Security / Offensive Cybersecurity / Generative AI
Deliverable
production ML models | research | client delivery
Required skills
Python programming, exploit development, vulnerability research, penetration testing, red teaming, adversarial testing, controlled experiment design, telemetry analysis, security controls implementation, model evaluation framework usage.
Preferred skills
Experience with PyRIT, coding agents, tool-using models, autonomous agent frameworks, containerized sandboxed execution, cloud-based test infrastructure.
Technologies
Python, PyRIT, LLMs, Generative AI systems, AI agents, model APIs, containers, cloud infrastructure.
Responsibilities
Execute controlled experiments using AI models to evaluate capabilities and vulnerabilities; analyze model trajectories to distinguish genuine capability from artifacts; develop and maintain cyber ranges and evaluation harnesses; partner with red-team operators to translate findings into security insights; document results through reports and briefings.
Seniority
Senior, hands-on IC