Staff Research Engineer: Scalable Multi-Agent Systems

Anthropic; New York (NY); On-site; USD 500,000 - 850,000; Full time

Job summary

Anthropic is seeking a generalist researcher-engineer to design and run large experiments at the intersection of research and engineering. You will build scalable systems for large agent teams and interpret surprising results to guide decisions.

You will own experiments end-to-end, collaborate with researchers and engineers, and communicate findings clearly while focusing on safe, beneficial AI and rigorous data analysis.

Qualifications

• Have significant software engineering, ML or research engineering experience.
• Have owned something substantial end to end, such as a large system, an evaluation or benchmark, a product, or a research project.
• Genuinely enjoy both research and engineering work.
• Think quantitatively about complex systems and verify numbers critically.
• Can work from a vague question rather than a spec.

Responsibilities

• Design, run and interpret large-scale experiments on agent teams with rigorous data analysis.
• Investigate how performance and efficiency change with team size, compute and task horizon.
• Build and scale systems that run very large agent teams reliably and debugs failures at scale.
• Design evaluations for long-horizon problems and keep results trustworthy.
• Build tooling and metrics to let researchers see what a large agent team is doing and why.
• Partner with research teams to run experiments on the platform and communicate findings clearly.

Skills

Software engineering

Machine learning

Research engineering

End-to-end ownership

Quantitative thinking

Communication

Impact motivation

Societal impact awareness

Education

Bachelor's degree or equivalent

Job description

Anthropic is seeking a generalist researcher-engineer to design and run large experiments at the intersection of research and engineering. You will build scalable systems for large agent teams and interpret surprising results to guide decisions.

You will own experiments end-to-end, collaborate with researchers and engineers, and communicate findings clearly while focusing on safe, beneficial AI and rigorous data analysis.