Latest AI JobsBrowse Jobs
Anthropic logo

Research Engineer, Code RL (Reinforcement Learning)

AnthropicSan Francisco, CA | New York City, NYAI Research & Engineering
Apply on company site
Salary
Not listed
Job type
Full-time
Posted
11 Jun 2026
Status
Open · checked 7d ago

About this role

The Research Engineer, Code RL position at Anthropic involves advancing artificial intelligence capabilities for software engineering through reinforcement learning in San Francisco or New York City. This role suits professionals with strong Python and reinforcement learning expertise who want to build safe and beneficial AI coding systems.

Role Overview

Anthropic is hiring a Research Engineer for the Code RL team within the Reinforcement Learning organization. In this capacity, the professional will enhance model abilities to write, test, debug, and deploy actual software across real codebases safely and efficiently. The position combines experimental research with engineering implementation, task design, reward signal creation, and pipeline optimization.

Key Responsibilities

Design reinforcement learning environments and specialized coding tasks for large language models.
Build verifiers and reward signals that accurately define and measure code quality.
Execute training experiments on frontier models to evaluate performance improvements.
Diagnose model behaviors during software engineering tasks to refine training pipelines.
Improve the overall reliability, speed, and performance of training infrastructure.
Collaborate closely with alignment, frontier red teams, and applied production training groups.

Requirements

Strong software engineering skills with deep Python expertise, including async and concurrent programming.
Ability to own end-to-end systems and troubleshoot issues across the entire technical stack.
Capability to balance research exploration with hands-on engineering execution.
Strong dedication to code quality, performance testing, and system reliability.
A minimum of a Bachelor's degree in a relevant field of study, or an equivalent combination of education, training, and experience.
Compliance with the company's location-based hybrid policy, requiring staff to be in the designated office at least 25% of the time.

Nice to Have

Prior experience working with reinforcement learning, RLHF, post-training, or large language model finetuning.
Background in building coding agents, execution sandboxes, evaluation harnesses, or developer tooling.
Familiarity with program analysis, verification methods, compilers, or formal testing.
Practical experience using PyTorch, large-scale distributed training, and machine learning performance profiling.
Direct experience with CUDA, GPU, or TPU kernels and accelerator performance.
Experience dealing with virtualization and sandboxed code execution platforms.

Skills and Tools

Python
PyTorch
Large Language Models (LLMs)
Reinforcement Learning
Distributed Training
CUDA and GPU Acceleration
Async Programming

Salary and Benefits

The annual base salary for this position ranges from $500,000 to $850,000 USD. Anthropic provides competitive compensation, optional equity donation matching, generous vacation allowances, parental leave, flexible working hours, and visa sponsorship support where applicable.

About the Company

Anthropic is a public benefit corporation focused on creating reliable, interpretable, and steerable artificial intelligence systems that benefit society. Operating as a collaborative and cohesive team, the organization treats AI research as an empirical science akin to physics and biology, striving for safe and high-impact long-term developments.

How to Apply

Interested candidates should visit the official Anthropic careers page to submit an application and review open positions safely.

Skills and tech stack

PythonPyTorchLLM

Questions candidates ask

What locations are available for this job?

This position is based in San Francisco, California or New York City, New York, operating under a location-based hybrid policy requiring at least 25% time in the office.

What is the salary range for the Research Engineer, Code RL role?

The annual salary range is between $500,000 and $850,000 USD.

Does Anthropic offer visa sponsorship for this position?

Yes, Anthropic sponsors visas for eligible roles and candidates, providing assistance through an immigration lawyer if an offer is extended.

What qualifications are required regarding education?

Candidates must hold at least a Bachelor's degree or demonstrate an equivalent combination of education, training, and experience in a relevant field of study.

What core technical skills are necessary for the role?

Applicants need strong software engineering skills, deep Python expertise including async programming, and familiarity with reinforcement learning or large language models.

Ready to apply?

The application is on the official Anthropic careers page.

Apply on company site