Staff+ Research Scientist, Multi-Agent

The role

This researcher will create and optimize environments and data for model training to maximize Claude's performance on agentic tasks. They will design agent harness configurations and implement quantitative benchmarks for large scale agentic tasks.

What you will do

• Help create and optimize environments and data for model training
• Ideate, develop, and compare the performance of different agent harness configurations
• Design and implement rigorous quantitative benchmarks for large scale agentic tasks
• Work with the product org to find solutions to challenges in applying agents to products
• Design and build reinforcement learning environments to train groups of Claudes
• Design and build agent affordances that unlock new capabilities and scales of agents
• Design and build a novel eval that measures how large teams of agents interact in groups

Required qualifications

• Bachelor's degree or an equivalent combination of education, training, and/or experience
• A field relevant to the role as demonstrated through coursework, training, or professional experience
• Experience with large-scale RL on language models
• Experience training multi-agent systems
• Good communication skills

Nice to have

• Enjoy going deeply into the roots of a problem and understanding its foundations
• An interest in working with other researchers on difficult tasks
• A passion for making powerful technology safe and societally beneficial
• Excitement for a mission-driven org with fast-paced, impactful work

Benefits

• Competitive compensation and benefits
• Optional equity donation matching
• Generous vacation and parental leave
• Flexible working hours
• Office space

Before you apply

• We do sponsor visas, but we aren't able to successfully sponsor visas for every role and every candidate

About the company

Anthropic is a public benefit corporation headquartered in San Francisco. Their mission is to create reliable, interpretable, and steerable AI systems.