Research Engineer, Takeoff Intel

Build and run evaluations measuring how AI systems assist Anthropic's own model development and capability acceleration.
- Design, build, and run capability evaluations and measurement instruments at scale; prototype fast and validate what to keep.
- Build data and analysis pipelines transforming large model outputs and telemetry into reliable metrics for research priorities.
- Review AI-written code as part of normal workflow; work closely with research scientists to define what's worth measuring.
- Shipped an evaluation, data product, or research library end-to-end; comfortable prototyping and discarding code rapidly.
- Experience running experiments on large language models and handling messy, large-volume data without over-engineering.

Anthropic is a frontier AI research and product company, with teams working on alignment, policy, and security. We post specific opportunities at Anthropic that we think may be high impact. We do not necessarily recommend working at other positions at Anthropic. You can read concerns about doing harm by working at a frontier AI company in our career review on the topic.