Security Engineer / Research Engineer, RL Security

Anthropic's mission is to create reliable, interpretable, and steerable AI systems. You will work on keeping the reinforcement learning (RL) system secure as it scales.

As an engineer in RL Engineering, you will work on the RL codebase and infrastructure, alongside engineers and researchers. The system is large and changes rapidly, so security work must keep up with that pace without impacting performance or velocity.

Key Responsibilities:

• Design, build, and operate security controls for RL training systems and infrastructure
• Work across the stack (orchestration, environments, training, inference) as needed
• Review designs and code, find security problems, and fix them
• Build shared libraries and tooling for other RL engineers and researchers
• Develop observability to understand system behavior
• Collaborate with researchers and performance engineers to ensure security changes preserve training correctness and performance
• Write clear design documents

Minimum Qualifications:

• Strong software engineering skills in Python and a systems language (Rust, C++, or Go)
• Experience designing, building, and operating large-scale distributed systems or security systems
• Strong understanding of security fundamentals
• Working knowledge of Linux, containers, and cloud infrastructure
• Experience debugging complex problems across multiple hosts and services
• Ability to reason about tradeoffs between security, performance, and reliability
• Strong written communication

Preferred Qualifications:

• Experience running ML training or inference infrastructure at scale
• Experience with infrastructure security (isolation, access control, networking, identity)
• Experience with container orchestration (Kubernetes) and sandboxed code execution
• Experience building observability for large fleets
• Familiarity with reinforcement learning or large language model training workloads

The annual compensation range for this role is $320,000-$405,000 USD.