jobs

Research Engineer, Interpretability

About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers,

Company

Anthropic

Location

San Francisco, CA

Status

Open

Posted

2025-11-07T22:35:29+00:00

Anthropic's Interpretability team is hiring a research engineer to build the training and inference infrastructure behind its mechanistic interpretability work — instrumented forward and backward passes, activation extraction, steering vectors and petabyte-scale activation pipelines. The job spans profiling and fixing bottlenecks across GPU or TPU stacks, designing tooling so researchers can experiment without engineering friction, and supporting production safety audits on frontier models. Candidates need 5–10+ years building software, strong Python, and either Rust, Go or Java. No interpretability research background is required.