THE OPPORTUNITY
A role worth
a closer look.
Test how frontier models perform on security-sensitive tasks and how reliably their safeguards resist misuse. Build evaluation tooling and abuse-detection probes, investigate adversarial behaviour, and work with policy and engineering colleagues to turn findings into better protection before and after model releases.
Experience highlighted by the employer
- Experience delivering software or ML evaluations alongside practical cybersecurity work.
- Python proficiency and clear communication of technical findings to different stakeholders.
Location & eligibility
US role with San Francisco and Washington, DC listed. The employer describes remote-friendly work but also states a general minimum 25% office policy. Visa sponsorship is considered case by case. Security clearance or eligibility is preferred, not a stated minimum requirement.
Compensation
USD $300,000–$405,000 / year. Employer-published annual salary range. Confirm the current offer with Anthropic.
Before you apply
This is TaskBull’s concise summary, not the employer’s full job description. Review the complete responsibilities, terms, accessibility information and privacy notice on the original application page.
Read the full employer listing ↗