Conceptual Reasoning Fellow

Details

Anthropic is seeking Fellows from philosophy, cognitive science, economics, policy, and law to work with research teams for four months. Fellows will combine deep domain expertise with frontier Claude models and internal tooling to evaluate, improve, and apply Claude’s capabilities to complex, judgment-intensive problems.

Program Overview

The fellowship includes a scoped project matched to the Fellow’s background, mentorship from Anthropic researchers, access to frontier Claude models and internal evaluation tooling, and collaboration with an incoming cohort and Anthropic’s broader research community.

Responsibilities

• Identify key open problems in your field, assess potential approaches, and use Claude to make progress on them.
• Identify major societal challenges where expertise in your field is important and assess Claude’s strengths and weaknesses in addressing them.
• Design evaluations and reinforcement learning environments that systematically test models in your field.
• Turn domain expertise into working tests of Claude’s reasoning capabilities.
• Collaborate with other Fellows to build environments that challenge Claude with difficult, judgment-heavy problems.

Requirements

• Excitement about the potential for AI to reason well about difficult questions where expert judgment matters.
• Hands-on experience using Claude and Claude Code in a day-to-day workflow.
• A PhD, or ongoing PhD studies, in philosophy, cognitive science, economics, policy, or law, or equivalent research experience.
• Ability to work in a fast-paced, high-agency environment and deliver concrete results over a few months.
• Existing authorization to work in the United States for the duration of the program.

Preferred Qualifications

• Experience designing evaluations, benchmarks, datasets, or expert rubrics.
• Demonstrated expertise through research, scholarship, or expert-level analytical writing.

Representative Projects

• Building difficult normative-judgment cases in ethics or formal epistemology.
• Designing experimental-design and model-comparison tasks.
• Writing economic modeling and identification problems with ground truth.
• Drafting regulatory and cost-benefit analysis tasks with expert rubrics.
• Building law datasets relevant to AI policy.

Compensation

The compensation is $100,000–$167,000 USD for the four-month fellowship, equivalent to $300,000–$500,000 USD annualized, plus other benefits.

Logistics

• The program runs from January 12 through May 8, 2027, and is full-time for the duration. Some fellowships may be considered for an extension.
• Fellows must attend a one-week in-person orientation in San Francisco.
• The default location is San Francisco, with remote engagements within the United States considered in certain circumstances.
• This is a fixed-term contract position. Fellows are engaged through an employer of record rather than employed directly by Anthropic.
• Anthropic is not currently able to sponsor visas for Fellows. Participants must have or independently obtain full-time work authorization in the United States.