Jobs

Anthropic is seeking a Research Engineer for their Interpretability team in San Francisco, CA, to reverse-engineer how trained models work and make advanced AI systems safe. The role focuses on mechanistic interpretability, aiming to discover how neural network parameters map to meaningful algorithms. Candidates should have 5-10+ years of software building experience, proficiency in languages like Python, and experience in empirical AI research. This position involves implementing and analyzing research experiments, optimizing workflows, and building tools to improve model safety.