Frontier AI Reasoning Fellow (PhD‑Level)

Anthropic Limited; Greater London; Hybrid; GBP 75,000 - 125,000; Full time

Benefits offered by this job

Mentorship from Anthropic researchers

Access to Claude models

Job summary

Anthropic Limited is offering a four-month research fellowship focused on AI safety, reasoning, and evaluation design. Fellows will work with Anthropic researchers, access Claude models, and collaborate with a cohort of peers to study hard questions in ethics, economics, and policy.

The program runs January 12 to May 8, 2027, with possible extension and remote options within the United States and United Kingdom in certain circumstances.

Qualifications

• Hold (or be completing) a PhD in philosophy, cognitive science, economics, policy or law, or have equivalent research experience.
• Hands-on experience using Claude and Claude Code in day-to-day workflow.
• Ability to work in a fast-paced, high-agency atmosphere for several months and deliver concrete results.

Responsibilities

• Identify key open problems in your field and use Claude to make progress.
• Assess societal challenges where your field’s expertise is crucial for addressing them.
• Design evaluations and RL environments that test models in your field.
• Collaborate with Fellows to build environments pushing Claude on hard, judgment-heavy problems.

Skills

Claude experience

Analytical writing

Academic research

Education

PhD in philosophy

PhD in cognitive science

PhD in economics

PhD in policy or law

Bachelor's degree or equivalent

Tools

Claude Code

Claude models

Job description

Anthropic Limited is offering a four-month research fellowship focused on AI safety, reasoning, and evaluation design. Fellows will work with Anthropic researchers, access Claude models, and collaborate with a cohort of peers to study hard questions in ethics, economics, and policy.

The program runs January 12 to May 8, 2027, with possible extension and remote options within the United States and United Kingdom in certain circumstances.