Research Engineer, Universes

About this role

Anthropic seeks a Research Engineer to develop advanced training environments for agentic AI systems. You'll blend research innovation with engineering implementation, creating realistic simulations and evaluations that enable AI models to handle complex, long-horizon tasks safely and effectively.

What you'll do

• Design and build next-generation agentic training environments
• Develop rigorous evaluations that measure genuine AI capability
• Collaborate across research and infrastructure teams to deploy environments into production
• Debug and iterate on research and production ML systems
• Contribute to research direction through technical discussions

What they're looking for

• Software engineering and robust infrastructure building
• Reinforcement learning and training methodologies
• Machine learning model training or evaluation
• RL environment or simulation system design
• Distributed systems or ML infrastructure experience
• Problem-solving with high technical judgment
• Ability to balance research exploration with implementation

Benefits

• Competitive salary: $500,000–$850,000 USD annually
• Remote-friendly with offices in San Francisco, Seattle, and New York
• Minimum 25% in-office presence required
• Visa sponsorship available
• Collaborative research culture with pair programming