Staff+ Site Reliability Engineer, Safeguards ML Infra

Overview

Staff+ Site Reliability Engineer for Safeguards ML Infra, owning production safeguards infrastructure that powers Claude’s safety systems and ensuring safe, verifiable rollouts during model launches.

What you'll do

• Stand up, configure, and verify safeguards for every new model release and act as safeguards point of contact during launch windows.
• Own off-cycle deployment of new safety classifiers from research, including canarying and post-deploy validations.
• Verify safeguards are provably live on the correct models across deployment platforms and detect/eliminate configuration drift.
• Turn launch runbooks and manual checks into continuous validation and repeatable deployment pipelines.
• Maintain a safeguards registry with provenance (what runs, on which model/platform, and who/when deployed).
• Participate in on-call and operational-duty rotations for service incidents and time-sensitive launches.

What you'll need

• Owned production change management at scale (deploy pipelines and config management systems).
• Experience with canary analysis and strong opinions about what “verified” means.
• Meaningful on-call experience for production systems, including incident response and postmortem-driven improvements.
• Hands-on experience deploying/operating cloud platforms at scale (AWS, GCP).
• Proficiency in Python (Rust experience is a plus, not required).
• Strong production judgment demonstrated through safely shipping changes to critical systems and automating away prior work.

Details

• Work arrangement: Remote-Friendly (Travel-Required) and expected to be in an office at least 25% of the time.
• Locations listed: San Francisco, CA; Seattle, WA; New York City, NY.
• Annual compensation range: $320,000–$485,000 USD.