Tech Stack
Tag name is followed by "@" symbol and proficiency level value.
About proficiency levels:
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
AI
API @ 4
Audit
LLM
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
Details
Anthropic is building reliable, interpretable, and steerable AI systems. As the company grows, internal software must support increasingly complex workflows. This role focuses on building Claude-first internal products that automate repetitive, multi-step work while preserving appropriate human decision-making.
The Staff+ full-stack engineer will build products across the stack, including user-facing applications and agents, while working directly with the teams that rely on these systems. Many workflows write to business-critical data, so correctness, data integrity, and reliability are essential.
Responsibilities
- Design and ship internal products end to end, from data models and APIs to user interfaces, for workflows used across Anthropic.
- Use Claude for tasks such as gathering decision-making context, drafting the next step in a process, and keeping records consistent across existing systems.
- Determine when an agent should act independently and when a well-designed user interface or human confirmation is more appropriate.
- Make agents safe to trust through scoped data access, guardrails, evaluations, audit trails, and clear human approval points.
- Build shared capabilities exposed through APIs and MCP tools so that people and agents use the same rules and approvals.
- Work with product users to identify workflow bottlenecks and improve speed and polish.
- Design for correctness in workflows involving business-critical data, including appropriate use of idempotency, reconciliation, and audit trails.
- Integrate with third-party systems and build the layer where people and agents complete their work.
- Set technical direction for a product area across multiple quarters.
- Lead engineers across teams through large, ambiguous projects and provide design reviews, code reviews, and mentorship.
- Operate shipped systems, own workflow reliability, and participate in on-call responsibilities.
Requirements
Minimum Qualifications
- Experience shipping production software as a backend-leaning full-stack engineer.
- Ability to design data models, APIs, and user interfaces.
- Experience building software on large language models or agents, with a clear perspective on when agents should act independently versus request approval.
- Experience leading the design and delivery of complex, user-facing full-stack products while requirements evolve.
- Experience designing for correctness and data integrity in systems where errors are costly and difficult to reverse.
- Experience working closely with non-engineering partners to translate ambiguous needs into clear requirements and reach decisions collaboratively.
- A bachelor's degree or equivalent combination of education, training, and experience. The required field of study must be relevant to the role through coursework, training, or professional experience.
Preferred Qualifications
- Experience building internal tools or platforms used by other teams.
- Experience taking an agent into production for a high-stakes workflow, including the evaluations and guardrails required for safe deployment.
- Experience building a zero-to-one product or rebuilding a product on which users depend.
- Experience building on third-party platforms through their APIs and deciding when to extend an existing system versus building a new one.
Compensation
- Annual salary: $405,000–$485,000 USD
Work Arrangement and Logistics
- Remote-friendly with travel required.
- Employees are currently expected to work from one of Anthropic's offices at least 25% of the time; some roles may require more office time.
- Offices for this role are in San Francisco, Seattle, and New York City.
- Anthropic sponsors visas for this role when possible and makes reasonable efforts to obtain a visa for candidates who receive an offer, with support from an immigration lawyer.
Benefits
Anthropic offers competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and office space for collaboration.