Staff Software Engineer, Safeguards Review Tooling

Anthropic; San Francisco (CA); On-site; USD 320,000 - 485,000; Full time

Job summary

Anthropic seeks an engineer for its Safeguards Review Tooling team in San Francisco to own the tools used by safety investigators and the platform beneath them. You will build analytics, privacy-preserving primitives, and a sandbox environment to accelerate Claude‑assisted review workflows.

This foundational role partners with policy, operations, data science, legal, and privacy teams to ensure effective, accurate, and trustworthy enforcement systems at scale.

Qualifications

• Technical background in full‑stack or platform engineering with architecture discussions.
• Experience shipping internal tools for operational users and improving workflows.
• Cross-functional collaboration with non-engineering teams such as policy or legal.
• Excellent communication to explain tradeoffs to non-technical stakeholders.
• Care about societal impacts of AI and safer systems.

Responsibilities

• Build investigation, review, and enforcement tooling for first‑ and third‑party surfaces.
• Develop the platform layer of reusable APIs, data storage, and backend services.
• Scale review through automation, including Claude-assisted workflows.
• Partner with policy, operations, legal, privacy, and data science teams.
• Build guardrails: permissions, audit trails, data-access controls, reviewer wellbeing features.
• Instrument tooling with metrics and align with privacy primitives and data retention commitments.

Skills

Full-stack engineering

Internal tooling

Cross-functional collaboration

Excellent communication

Societal impact

Education

Bachelor’s degree in a relevant field

Tools

Review tooling platforms

Claude/LLM integration

Job description

Anthropic seeks an engineer for its Safeguards Review Tooling team in San Francisco to own the tools used by safety investigators and the platform beneath them. You will build analytics, privacy-preserving primitives, and a sandbox environment to accelerate Claude‑assisted review workflows.

This foundational role partners with policy, operations, data science, legal, and privacy teams to ensure effective, accurate, and trustworthy enforcement systems at scale.