Staff+ Fullstack Software Engineer, Safeguards Engineering

San Francisco, CA | New York City, NY · San Francisco · USD 405,000-485,000 / year · 8+ years experience · Hybrid · Posted Oct 9, 2026

At a glance

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.
Note: this job post is for Fullstack Software Engineers only. If you are a Backend Engineer, please see our Backend Software Engineer post here.

The Safeguards team builds safety and oversight mechanisms to monitor AI models, prevent misuse, and ensure user well-being.
This role focuses on developing systems to detect unwanted model behaviors and enforce acceptable use policies.
Engineers will work across the stack to build internal review tools and human-in-the-loop supervision interfaces.

Design and build internal review tools including dashboards, case queues, and evidence viewers.
Build interfaces for human supervision of agents, including streaming output, interruption, and approval flows.
Surface abuse patterns to research teams through visualizations and exploration tools.
Own the frontend architecture for tools handling sensitive data at scale.
Ensure performance under large datasets and correctness under real-time updates.
Implement end-to-end access controls for internal tooling.

Bachelor’s degree in Computer Science, Software Engineering, or comparable experience.
Proficiency in TypeScript and modern React.
Comfortable in Python for backend work.
Ability to work across the stack.
Strong communication skills and ability to explain complex technical concepts to non-technical stakeholders.

8+ years of experience in a software engineering position.
Experience building data-dense internal tools or operator interfaces.
Experience with integrity, spam, fraud, or abuse detection and mitigation.
Experience building trust and safety detection mechanisms and intervention for AI/ML systems.
Experience with prompt engineering, jailbreak attacks, and other adversarial inputs.
Experience working closely with operational teams to build custom internal tooling.

Location-based hybrid policy requires staff to be in the office at least 25% of the time.
Some roles may require more time in the office.