
Anthropic Research Resident
AI
Anthropic is seeking Research Residents for a paid, fixed-term program of six weeks to six months to work on empirical AI safety and security research alongside Anthropic researchers. Candidates should have led ambitious work in technical fields such as academia, technical founding, or staff-level engineering, with a concrete scoped project, mentorship, and access to Claude models and GPU compute.
What you'll do
- Work on a concrete, scoped research project matched to your background
- Receive direct mentorship from Anthropic researchers
- Conduct empirical research questions in collaboration with Anthropic's research teams
What you need
Must-have
Machine LearningNLPAI SafetyLLMsReinforcement LearningKubernetesEmpirical AI Research
Nice-to-have
Research paper authoringAdversarial Machine LearningAI SecurityInterpretabilityScalable OversightModel OrganismsAI WelfareCybersecurity
Soft skills
CommunicationCollaborationFast-paced adaptabilityImplementing ideas quickly
AI ResearchAI SafetyAlignmentInterpretabilityMachine LearningResearch ResidencyAnthropicLondonRemoteFixed-term