What it offers
The Anthropic Residency offers established technical leaders—such as quantitative professors, technical founders, and staff-level engineers—an opportunity to transition into full-time research on AI safety and alignment. Residents work on core empirical research topics alongside Anthropic researchers in areas including AI safety and alignment, scalable oversight, adversarial robustness, model interpretability, AI security, and ML systems. The program provides one-on-one mentorship, structured self-study time, talks, and direct access to dedicated compute resources. Programme duration: 6 weeks to 6 months.
Placement and commitment
Understand the appointment, placement structure, cohort expectations, and commitments required before and after the fellowship.
Who Can Apply
Open to experienced technical leaders (such as professors in quantitative disciplines, technical founders, or staff-level engineers) who are motivated to transition into full-time AI safety research. Prior background in machine learning, a PhD, or published papers are not required.
Open to applicants from
Not specified
Education levels
Not specified
Age and gender
All
Not sure you qualify? MorrowGuide checks each rule against your profile and shows what is still unknown.
Key Dates
Application Deadline
What’s included
Residents receive competitive salary compensation equivalent to $350,000 to $500,000 per year based on seniority. Benefits include one-on-one guidance from an Anthropic mentor, structured self-study support, compute budget for research projects, and flexible placement start dates and durations.
Application requirements
Interested candidates must complete an online application via either the US or UK posting by November 22, 2026, indicating their preferred duration (six weeks, three months, or six months).