Staff Software Engineer, Internal Products

About the role

As Anthropic grows, more of the work our own teams do every day runs through internal software, and that software has to keep up. We are building that software, and we are building it Claude-first: Claude should take on as much of the repetitive, multi-step work as it safely can, and people decide what needs a human.

We’re looking for Staff+ full-stack engineers to drive this work. You’ll build across the stack: the products people use and the agents that work alongside them. You’ll work directly with the teams who rely on what you build. Many of these workflows write to business-critical data, so correctness matters as much as speed.

This is a greenfield opportunity: you’ll shape the underlying systems, ship agents that take real work off people’s plates, and design how people and agents get the right context to make good decisions.

Key responsibilities

• Design and ship internal products end to end, from the data model to the UI, for workflows that teams across Anthropic depend on.

• Put Claude to work on real tasks, such as gathering the context a decision needs, drafting the next step of a multi-step process or keeping records consistent across the systems teams already use, and decide where an agent should act and where a well-designed UI is the better answer.

• Make agents safe to trust with real work: scoped access to data, guardrails, evals, audit trails, and clear points where a person confirms before anything happens.

• Build one set of shared capabilities that people and agents both use, exposed as APIs and MCP tools, so the same rules and approvals apply to whoever makes the call.

• Treat speed and polish as product requirements. Spend time with the people who use what you build, find where their work slows down, and remove those steps.

• Build for correctness. When a workflow writes to business-critical data, a mistake travels into other systems, so you’ll know where idempotency, reconciliation and audit trails matter and where they don’t.

• Integrate with the third-party systems teams already rely on, and build the layer on top of them where people and agents get their work done.

• Set technical direction for a product area over multiple quarters, lead engineers across teams through large, ambiguous projects, and make the engineers around you better through design reviews, code review and mentorship.

• Operate what you ship: own the reliability of the workflows people depend on and be on call for them.

Minimum qualifications

• Experience shipping production software as a backend-leaning full-stack engineer, comfortable designing a data model, an API and the UI on top of them.

• Experience building software on LLMs or agents, and a point of view on when an agent should act on its own and when it should ask first.

• Experience leading the design and delivery of complex, user-facing products across the full stack, through requirements that shifted while you built.

• Experience designing for correctness and data integrity from the start, in systems where errors are costly and hard to undo.

• Experience working closely with non-engineering partners, turning ambiguous needs into clear requirements and reaching decisions together.

Preferred qualifications

• Experience building internal tools or platforms that other teams rely on. A plus, not a requirement; strong product generalists are just as welcome.

• Experience taking an agent into production for a high-stakes workflow, including the evals and guardrails that made it safe to ship.

• Experience building a zero-to-one product, or rebuilding one that people already depended on every day.

• Experience building on top of third-party platforms through their APIs, and knowing when to extend a system and when to build your own.