Staff ML Systems Engineer - Safeguards Path

Socket.dev; San Francisco (CA); Hybrid; USD 320,000 - 485,000; Full time

Job summary

Anthropic is hiring for a role in the Safeguards ML Sampling Path team that builds and operates production services powering Claude's safety systems. You will design, implement, and run backend systems handling every token on the generation path, with a focus on latency, reliability, and cost efficiency.

The role requires strong distributed systems expertise, experience with high QPS production services, and the ability to drive design decisions across teams.

Qualifications

• Bachelor's degree or equivalent practical experience.
• 8+ years of software engineering experience with production systems.
• Familiarity with large language model inference systems is a plus.

Responsibilities

• Design, build, and operate backend systems that process every token on the generation path for Claude requests.
• Own latency and reliability end-to-end: define and maintain SLOs and error budgets; lead incident response and postmortems.
• Ship changes rapidly and safely on the hot path with canarying, rollouts, and latency gating; optimize per-token performance as traffic grows.
• Set technical direction for the sampling path; lead design reviews; mentor engineers; raise operational bar for Safeguards.

Skills

8+ years industry software engineering

LLM inference familiarity

Education

Bachelor's degree

Job description

Anthropic is hiring for a role in the Safeguards ML Sampling Path team that builds and operates production services powering Claude's safety systems. You will design, implement, and run backend systems handling every token on the generation path, with a focus on latency, reliability, and cost efficiency.

The role requires strong distributed systems expertise, experience with high QPS production services, and the ability to drive design decisions across teams.