About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of…
Security Data Engineer at Anthropic owning data systems for hardware inventory, chain of custody, and continuous control monitoring across data centers. Builds pipelines, AI agents, and field software to automate security posture calculations and compliance evidence, using SQL, Python, and AI-assisted development.
Staff Security Engineer in Anthropic's Security Labs building high-risk security prototypes (isolated clusters, cryptographic verification, hypervisor isolation, formal verification) for AI systems using Python/Rust/Go/C++ in a hybrid role across SF/NYC/Seattle.
Foundational privacy engineering role establishing privacy-by-design for Anthropic's frontier AI systems (Claude). Architect privacy-preserving training/inference at scale using differential privacy, federated learning, and secure computation; build data governance infrastructure; translate GDPR/CCPA/HIPAA into automated controls; lead privacy reviews across research, product, and engineering. Cor
Cookies?
A session cookie keeps you signed in. If you accept, we also use analytics cookies (Google Analytics and PostHog) to understand usage. Learn more
We couldn't check your fit for this role — add a CV to your profile to see it next time.