Technical Program Manager, Agent Safety & Security

Location: San Francisco, CA or New York City, NY; hybrid policy requiring staff to work from one of the offices at least 25% of the time, with some roles requiring more office time.

Annual salary: $365,000–$435,000 USD.

Company

Anthropic is a public benefit corporation building reliable, interpretable, and steerable AI systems designed to be safe and beneficial for users and society.

What you will do

• Maintain and extend the company-wide roadmap for agent safety and security work, ensuring accuracy and relevance as capabilities and threats evolve.
• Translate program goals into measurable phase exit criteria, metrics, and dashboards.
• Run leads meetings, working groups, design reviews, and executive check-ins; record decisions, distribute situation reports, and follow up on open issues.
• Prepare executive reports, design and decision documents, company-wide communications, and customer-facing updates.
• Maintain the risk register, track exceptions and risk acceptances, and escalate difficult decisions to the appropriate owners.
• Coordinate TPMs and engineering managers across Security, Safeguards, Research, Privacy, Legal, and Product; organize control-testing exercises and address cross-team staffing and ownership gaps.

Requirements

• 10+ years of experience in agent or infrastructure security, ML classifiers and monitoring, evaluations, inference infrastructure, protection of sensitive organizational data, or shipping security controls to customers.
• Experience leading company-wide technical programs, especially security programs with multiple workstreams, executive sponsors, and other program managers reporting into the program.
• Strong background in at least one relevant program domain and the ability to quickly ramp up on the others.
• Experience working closely with ML research teams and understanding how models are trained and evaluated.
• Ability to work independently, set direction, make decisions, and escalate appropriately.
• Ability to hold senior engineers, researchers, and executives accountable without formal authority.
• Clear, fast writing for audiences ranging from executive briefings to company-wide announcements.
• Experience using AI tools such as Claude to automate tracking, reporting, and follow-up.
• Interest in AI safety and keeping powerful systems under human oversight.
• Bachelor’s degree or an equivalent combination of education, training, and experience in a field relevant to the role through coursework, training, or professional experience.

Culture & Benefits

• Collaborative research environment focused on large-scale efforts in steerable and trustworthy AI.
• Frequent research discussions and strong emphasis on communication and impact.
• Competitive compensation with optional equity donation matching.
• Generous vacation and parental leave, flexible working hours, and an office environment for collaboration.
• Visa sponsorship may be available; Anthropic makes reasonable efforts to support visas and provides access to an immigration lawyer.

Hiring process

• Applications are accepted on a rolling basis with no stated deadline.
• Candidates should follow Anthropic’s guidance on AI usage during the application process.