Staff / Principal Security Engineer, AI Safety and Security
Core
Lead security work at the boundary of AI safety and security for scientific discovery systems, investigating misuse risks and protecting sensitive models/data.
Role type
Staff/Principal Security Engineer (AI Safety)
Builds
Threat models, adversarial tests, security requirements, and evidence-based recommendations for safe deployment of AI systems.
Domain
AI Safety, Security, Scientific Discovery
Deliverable
production ML models
Required skills
Security architecture, offensive security, incident response, threat modeling, adversarial testing, technical investigation, clear writing, cross-disciplinary reasoning
Preferred skills
Securing ML infrastructure, evaluating safeguards/adversarial robustness, translating risk assessments to technical requirements, working in high-consequence scientific environments
Technologies
AI models, scientific workflows, automated laboratories
Responsibilities
Investigate how dangerous misuse or compromise could occur across models and services; Run focused adversarial tests and build prototypes to identify weaknesses; Turn capability assessments into concrete security requirements; Define evidence criteria to demonstrate controls work together; Prioritize security changes and verify results with engineering teams; Develop methods and guidance for others to apply to new problems
Seniority
Staff/Principal, hands-on IC with strategic impact
