Cyber Evaluations Engineer (AI)

Location: Remote-friendly within the United States, with locations in San Francisco, CA and Washington, DC. Staff are expected to work from one of the offices at least 25% of the time, and some roles may require more office attendance.

Annual salary: $300,000–$405,000 USD

Company

Anthropic builds reliable, interpretable, and steerable AI systems designed to be safe and beneficial.

What you will do

• Design and run capability, uplift, and safety evaluations for cyber-relevant risks in new AI models.
• Execute safeguard-robustness testing before major model releases and communicate findings to stakeholders.
• Design, prototype, and tune detection probes for cyber misuse, including jailbreaks and prompt bypasses.
• Partner with cyber policy and engineering teams to build layered abuse-detection architecture and measure its precision and coverage.
• Build and maintain internal tooling for running and scoring evaluations.
• Translate evaluation findings into improvements to model safeguards.

Requirements

• Experience building or running evaluations, benchmarks, or test suites for software or ML systems on fixed timelines.
• Hands-on cybersecurity experience, such as CTF participation, vulnerability research, exploit development, or security research.
• Proficiency in Python.
• Strong communication skills for presenting evaluation results to cross-functional and policy stakeholders.
• Bachelor’s degree or equivalent education, training, or professional experience in a relevant field.

Nice to have

• Offensive-security or security-research experience, including AI security benchmarks.
• Experience analyzing adversarial or abuse data and working with AI/ML evaluation frameworks.
• Experience authoring detection content such as Sigma, YARA, Suricata, or SIEM rules, or building ML-based abuse detection.
• Experience with coordinated vulnerability disclosure and pre-release testing under confidentiality constraints.
• Active secret security clearance or higher, or eligibility to obtain one.

Culture & Benefits

• Collaborative research environment focused on large-scale AI safety efforts.
• Flexible working hours and office collaboration.
• Generous vacation and parental leave.
• Competitive benefits and optional equity donation matching.
• Visa sponsorship is available, with immigration-lawyer support, although sponsorship cannot be guaranteed for every role or candidate.