Staff+ Software Engineer, Safeguards Infrastructure
Anthropic London £325,000 – £395,000 pa
Builds foundational infrastructure for AI safety safeguards, including data storage, metrics and evaluation systems, and tooling for human and agentic review. Involves maintaining real-time, multi-layered defenses against model misuse at scale while reducing manual oversight. Works across the stack, primarily in Python, with opportunities in TypeScript or Rust.