About the job

Join Anthropic’s Reinforcement Learning team to advance AI systems. You will work on reinforcement learning research, large language models, and agentic AI capable of tool use for open-ended tasks. Responsibilities include designing training environments, scaling RL infrastructure, developing safe and reliable AI systems, improving reasoning abilities, and prototyping internal tools. The role blends research and engineering, so proficiency in Python, async/concurrent programming, and machine learning frameworks (PyTorch, TensorFlow, JAX) is required. Strong candidates may have experience with reinforcement learning techniques, distributed systems, virtualization, and high-performance computing. Visa sponsorship is available for qualified applicants.