- San Francisco
- On-site
- Full-time · Mid
Lead research on securing frontier AI systems, designing experiments, developing defenses, and collaborating with Security and infrastructure teams.
Responsibilities
- Design and run experiments to discover failure modes in sandboxed AI environments
- Develop threat models and novel adversarial tests for multi‑boundary attacks
- Invent, prototype, and evaluate defenses to strengthen isolation
- Develop automated adversarial tests and repeatable evaluations
- Partner across Research, Security, and infrastructure teams to prioritize risks and deploy safeguards
Requirements
- Nice to have: operating systems
- Nice to have: containers
- Nice to have: virtualization
- Nice to have: distributed infrastructure
- Nice to have: AI agents
