Anthropic jobs

Staff Software Engineer, Node Infra at Anthropic focuses on the full lifecycle management of AI accelerator capacity—ingestion, provisioning, health checking, and automated repair across thousands to hundreds of thousands of compute nodes. You'll own technical strategy for node lifecycle management, design systems that detect and remediate unhealthy hardware automatically, and drive cross-team initiatives to build AI clusters across multiple clouds and accelerator families. Required: deep expertise in distributed systems and cloud platforms (Kubernetes, Terraform, AWS/GCP/Azure), strong proficiency in systems languages (Rust, Go, or Python), hands-on experience with ML accelerators (GPUs, TPUs, Trainium), and a track record leading complex multi-quarter technical initiatives across teams. Preferred qualifications include 8+ years of software engineering with technical leadership experience, large-scale compute infrastructure management (10K+ nodes), depth in Kubernetes internals or cluster orchestration, and low-level systems experience. The role offers competitive compensation ($320–405K annually) and sits at a critical intersection of Anthropic's AI research and infrastructure strategy. This is a senior technical leadership position requiring proven ability to shape infrastructure strategy and mentor engineers, with locations in San Francisco, NYC, or Seattle.