Product Manager, Safeguards (Child Safety)

Details

Anthropic is building AI assistants that are helpful, harmless, and honest. As usage grows, the Safeguards team works to protect users from risks of powerful AIs as well as ethical, technical, and social risks from generative AI. The Safeguards team builds protections for new AI features enabled by research teams and protects new products and surfaces developed by product teams.

As a Product Manager for the Safeguards team, you will own the ideation, design, development, and deployment of Safeguards systems and relevant product UX to ensure Anthropic’s frontier models are advanced safely to users across various cloud platforms. You will work closely with research and product teams to develop detections, evals, interventions, and tools to measure and mitigate deployment and user risks.

You are deeply committed to making AI safe and beneficial for humanity. You are aware of AI risks and committed to working with experts and coming up with ideas for Anthropic to implement. You have deep technical expertise in development, deployment, and measurement of Safeguards systems, and you thrive in rapidly moving and ambiguous environments.

Responsibilities

• Determine how to build in safety by design upstream and leverage downstream defenses for Anthropic’s frontier models, AI products, customers on different surfaces — Claude.ai, 1P API, and external Cloud providers.
• Write safety evals and communicate externally about safety.
• Drive impact via ruthless prioritization: clearly define problems, solution options, and requirements toward MVP vs. ideal state, with clarity on both business & technical tradeoffs.
• Align and collaborate with policy, enforcement, research, engineering, and cross-functional stakeholders.
• Understand the AI landscape and ecosystem to plan for mitigation of deployment risks of increasingly powerful models and determined adversaries.
• Lead the development of metrics to understand the area: performance, blindspots, and to inform future project planning.

Requirements

• Ability to make technical tradeoff decisions; experience working across policy experts, AI/ML research engineers, and software engineering teams to design and build safety systems.
• Strong user understanding of how Anthropic products are used, their Safeguards concerns, and how to provide the best solutions.
• Demonstrated ability to build product and engineering strategy across multiple cross-functional teams in a rapidly changing space.
• Demonstrated experience designing and building metrics to evaluate risks, system performance, user impact, and to make crisp tradeoffs.
• Very strong ability to navigate and prioritize amidst rapidly changing product specs, flex into different domains, bring clarity, and execute.
• Evidence of exercising judgment and decision making in ambiguous situations.
• Planning, building, launching, and measuring new products/systems in a zero-to-one environment.
• Ability to clearly articulate complex technical concepts to non-technical audiences in written and verbal communication.
• Think creatively about risks and benefits of new technologies beyond past checklists and playbooks.

Preferred Qualifications

• 5+ years in product management with a focus on fast problem understanding, building roadmaps with tractable progress, getting into details on data, detection & interventions, infrastructure & tools, and/or evals.

Compensation

Annual Salary: $305,000 - $385,000 USD.

Logistics

• Minimum education: Bachelor’s degree or an equivalent combination of education, training, and/or experience.
• Location-based hybrid policy: currently expects all staff to be in one of its offices at least 25% of the time (some roles may require more time in offices).
• Visa sponsorship: Anthropic sponsors visas; if an offer is made, it will make every reasonable effort to get a visa and retains an immigration lawyer to help.