AI Research Fellowship – 4‑Month Full‑Time Role

About the role

The Anthropic Fellows Program offers a four‑month, full‑time research fellowship focused on safe and beneficial AI. Fellows work on an empirical project aligned with Anthropic’s research priorities, aiming to produce a public output such as a conference paper.

Key responsibilities

• Design and execute an AI research project using open‑source models and public APIs.
• Collaborate closely with Anthropic mentors and receive direct guidance.
• Manage a compute budget of approximately $15k per month.
• Engage with the broader AI safety and security research community.
• Prepare and submit a research paper or comparable public deliverable.

Required profile

• Strong technical background and passion for AI safety research.
• Ability to work independently and meet weekly milestones.
• Motivation to contribute to socially beneficial AI systems.
• Underrepresented candidates are especially encouraged to apply.

Required skills

• Experience with open‑source AI models.
• Familiarity with public APIs for machine‑learning services.

What we offer

• Weekly stipend of $3,850 USD (or equivalent in GBP/CAD).
• Compute budget of ~$15,000 per month and other research expenses.
• Direct mentorship from Anthropic researchers.
• Access to a shared workspace in London (or Berkeley).
• Connection to a global AI safety research community.