Staff Software Engineer — AI Safety & Evaluations

Aicompliance; New York, San Francisco, Seattle (NY, CA, WA); Vor Ort; USD 320.000 - 485.000; Vollzeit

Zusammenfassung

Anthropic in New York City, NY; San Francisco, CA; Seattle, WA is hiring a Staff+ Software Engineer for Safeguards Evals to build evaluations and datasets focused on model safety, misuse, prompt injection, and agentic investigations. You will help integrate safety evaluations into training and release pipelines.

The role involves collaborating with research and product teams, improving evaluation tooling, and ensuring scalable, reliable safety checks across deployments.

Aufgaben

• Builds evaluations and datasets for model safety, misuse, prompt injection, and agentic investigations.
• Integrates safety evaluations into training and release pipelines.

Jobbeschreibung

Anthropic in New York City, NY; San Francisco, CA; Seattle, WA is hiring a Staff+ Software Engineer for Safeguards Evals to build evaluations and datasets focused on model safety, misuse, prompt injection, and agentic investigations. You will help integrate safety evaluations into training and release pipelines.

The role involves collaborating with research and product teams, improving evaluation tooling, and ensuring scalable, reliable safety checks across deployments.