AI Safety Researcher

Anthropic is seeking an AI Safety Researcher to work on some of the most important open problems in making AI systems safe and beneficial. You'll conduct original research on topics including RLHF, constitutional AI, scalable oversight, and robustness.

You'll have significant autonomy to pursue research directions you find most promising, while collaborating with a world-class team of researchers and engineers. Your work will directly influence the development of Claude and future AI systems.

• PhD in ML, CS, statistics, or related field (or equivalent research experience)
• Strong publication record in ML, AI safety, or adjacent fields
• Experience with large language models
• Excellent written and verbal communication
• Track record of independent research

• Research-focused environment with high autonomy
• Competitive compensation and equity
• Conference travel budget
• Compute resources for research
• Collaborative culture with top researchers