SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Anthropic is seeking a Staff Site Reliability Engineer to join the Safeguards ML Infrastructure team, which designs, builds, and operates the production infrastructure powering Claude's safety systems. This role sits at the center of operational work ensuring safeguards are properly configured and deployed for model launches.
You will own the off-cycle deployment of new safety classifiers, serve as launch captain for model releases, and ensure safeguards are correctly configured across all deployment platforms (1P, AWS Bedrock, GCP Vertex, etc.). Key responsibilities include canarying changes, verifying safeguards are live on the correct models, holding rollback authority, and leading incident response when issues arise.
A core focus is automation: converting launch runbooks into tooling, hand-built checks into continuous validation, and one-off deploys into repeatable pipelines. You'll build and maintain a safeguards registry with full provenance tracking what runs in production, on which model, on which platform, and deployment history. You'll participate in on-call rotations covering service incidents, model provisioning, and time-sensitive launches.
The ideal candidate has 8+ years of software engineering or SRE experience with deep expertise in production change management at scale—deploy pipelines, config management systems, rollout safety, and launch readiness. You should have run high-stakes releases as a launch captain or incident commander for systems where bad deploys have real consequences. Strong on-call experience with incident response and postmortem-driven improvements is essential. Hands-on cloud platform experience (AWS, GCP) at scale is required. Proficiency in Python is mandatory; Rust experience is a plus. Familiarity with LLM inference systems and transformer-based models is valuable but not required—you will learn on the job. What matters most is production judgment and a track record of automating yourself out of last quarter's work.
