Anthropic is looking for a Security Labs Engineer to lead experimental security projects that push the boundaries of AI safety. The role involves building prototypes, testing against real workloads, collaborating with research and infrastructure teams, and documenting outcomes to shape Anthropic’s security architecture. The position is based in San Francisco, CA and is onsite.
Security Labs Engineer at Anthropic
On-site - San Francisco, CA, United States
More jobs at AnthropicRequirements
Skills
- Strong software engineering fundamentals
- Expertise in firmware or hardware security
- Applied cryptography knowledge
- Experience with OS/kernel/hypervisor internals
- Formal methods and verification skills
- Reverse engineering and exploit development proficiency
- High-assurance / cross-domain systems experience
- Track record of building and shipping products independently
- Comfortable working across multiple domains and stack layers
Responsibilities
- Own Security Labs projects from concept to prototype to hand‑off or exit
- Scope bets, build prototypes, run them against real workloads
- Stand up novel security infrastructure rapidly (isolated clusters, attestation chains, hypervisor runtimes, verification tooling)
- Identify receiving teams early, collaborate, and hand over deliverables they can own
- Embed with research and infrastructure teams to test workflow survivability and document gaps
- Translate experimental results into short writeups that influence Anthropic’s long‑term security architecture
- Help select the next round of bets and shape industry best practices
Technologies
Cryptographic verification (zero‑knowledge proofs, attestation chains)Hypervisor‑isolated microVM substrateFormally verified ML kernel compilationSecurity infrastructure for isolated research clusters
See if your resume is ready for this job
See how our AI can optimize your resume and improve your chances for this role.