Staff+ Site Reliability Engineer, Safeguards ML Infra
About this role
Engineers on this team configure, verify, and roll out safety classifiers and safeguards across every platform where Claude runs, while also leading incident response and maintaining a safeguards registry with full provenance. The Safeguards ML Infra team designs, builds, and operates the production infrastructure powering Claude's safety systems. Team members ensure safeguards are properly configured for model launches, handle off-cycle deployments, and transform manual launch runbooks into automated, repeatable pipelines using tools like Claude.
This role exists to bring rigorous production change management and operational judgment to frontier AI safety systems. Ideal candidates possess deep experience managing high-stakes releases, handling complex on-call rotations, and eliminating operational toil through robust automation. Professionals who thrive in the critical path of deployment pipelines and enjoy turning manual checks into continuous validation will find themselves well-suited for this engineering challenge.
What you will do
- Act as launch captain to stand up, configure, and verify safeguards for every new model release while serving as the primary point of contact during release windows.
- Manage the off-cycle deployment of new safety classifiers by canarying rollouts, running post-deploy validations, and investigating discrepancies.
- Verify that correct safeguards are active on designated models across multiple deployment platforms and eliminate configuration drift.
- Convert manual launch runbooks and hand-built checks into continuous validation and repeatable deployment pipelines.
- Build and maintain a comprehensive safeguards registry tracking what is running in production, on which model, and on which platform.
- Participate in on-call and operational-duty rotations covering service incidents, model provisioning, and safety launches.
What you need
- Demonstrate extensive experience in production change management at scale including deploy pipelines and config management systems.
- Possess a track record of running high-stakes releases as a launch captain, incident commander, or release owner.
- Bring meaningful on-call experience for production systems including incident response and postmortem-driven improvements.
- Show a strong history of reducing operational toil through automation and transitioning teams to self-serve pipelines.
- Have hands-on experience deploying and operating on cloud platforms such as AWS and GCP at scale.
- Exhibit proficiency in programming with Python.
Nice to have
- Experience with Rust
- 8+ years of industry software engineering or site reliability engineering experience
- Experience running launch or production-readiness review processes across multiple teams
- Familiarity with LLM inference systems and the operational characteristics of transformer-based models
Skills and tech stack
Questions candidates ask
What is the remote work or office policy for this job?
Anthropic operates on a location-based hybrid policy where staff are expected to be in one of the offices at least 25% of the time, though some roles may require additional in-office presence.
What is the annual compensation range for the role?
The annual salary range for this position is $320,000 to $485,000 USD.
What programming languages and cloud platforms are required?
Engineers must be proficient in Python and have hands-on experience deploying and operating on cloud platforms like AWS and GCP at scale.
Is machine learning or transformer architecture experience required to apply?
Familiarity with ML research or transformer architectures is not required, as candidates will learn those concepts on the job.
Does the company offer visa sponsorship for this position?
Anthropic does sponsor visas and retains an immigration lawyer to assist candidates who receive an offer with the visa process.