Full-Stack Software Engineer, Reinforcement Learning

at Anthropic

📍 New York City, United States
📍 San Francisco, United States

USD 300,000-405,000 per year

MIDDLE

✅ Hybrid

✅ Visa Sponsorship

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.

API @ 3 AWS @ 3 Audit @ 3 CI/CD @ 3 Communication @ 3 Docker @ 3 GCP @ 3 LLM @ 2 Observability @ 3 Python @ 5 React @ 5 Reinforcement Learning TypeScript @ 5

Details

Build the platforms, tools, and interfaces that support reinforcement learning environment creation, data collection, and training observability at Anthropic. This role owns product surfaces end-to-end, from backend services and APIs to web interfaces used by researchers, external vendors, and data labelers.

Responsibilities

Build and extend web platforms for reinforcement learning environment creation, management, and quality review, including configuration, versioning, and validation workflows.
Develop vendor-facing interfaces and tooling for creating, submitting, and iterating on training environments.
Design and implement large-scale human data collection platforms with labeling workflows, quality assurance systems, and feedback mechanisms.
Build evaluation dashboards and observability UIs for monitoring environment quality, training run health, and reward hacking.
Create backend services and APIs connecting environment authoring tools, data collection systems, and reinforcement learning infrastructure.
Build scalable code data generation pipelines for programming tasks across languages and difficulty levels.
Develop onboarding automation and documentation tooling for vendors and internal users.
Partner with reinforcement learning researchers, data operations, and vendor management to translate ambiguous requirements into well-scoped products.

Requirements

Strong software engineering fundamentals and full-stack experience, including ownership from database schema through frontend.
Proficiency in Python and a modern web stack such as React and TypeScript.
A track record of shipping systems that solve difficult problems and improve team productivity.
High agency and the ability to drive work forward independently.
Strong UX judgment and the ability to build intuitive interfaces for technical researchers and non-technical labelers.
Clear communication skills and the ability to turn vague requirements into well-scoped work.
Bachelor’s degree or an equivalent combination of education, training, and experience.
Relevant education, training, or professional experience in a field related to the role.

Preferred Qualifications

Experience building data collection, labeling, or annotation platforms at scale.
Experience with multi-tenant platforms, role-based access, audit trails, and vendor management workflows.
Experience with GCP or AWS, Docker, and CI/CD pipelines.
Familiarity with LLM training, fine-tuning, or evaluation workflows.
Experience with asynchronous Python using Trio or asyncio, or with high-throughput API design.
Experience building dashboards, monitoring, or observability tooling.
Experience working with external vendors or partners on technical integrations.

Benefits

Anthropic offers competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and an office environment for collaboration. Staff are currently expected to work from an Anthropic office at least 25% of the time, with some roles requiring more office time.