Engineering Manager – Safeguards Interventions

About the role

Anthropic is seeking an Engineering Manager to lead the Safeguards Interventions team. This team designs, builds, and operates the systems that respond when a safety detector fires across all Anthropic products, including APIs, first‑party applications, and third‑party cloud deployments.

Key responsibilities

• Hands‑on leadership of a growing engineering team, owning roadmap, OKRs, and delivery.
• Coordinate cross‑functional work with ML Infra, Research, Product, Policy, Legal, and cloud partners.
• Define quality standards for interventions, backed by measurement, and communicate safety‑product trade‑offs to leadership and external stakeholders.
• Own production reliability: incident response, post‑mortems, SLOs, and verification processes to prevent repeat incidents.

Required profile

• Experience managing engineering teams that ship production ML or safety‑enforcement systems affecting users.
• Proven track record running high‑stakes, compliance‑adjacent production systems, including on‑call duties and regulator‑driven requirements.
• Strong focus on measurement and evaluation, with a history of deprecating ineffective systems.
• Ability to navigate ambiguous, multi‑stakeholder trade‑offs (safety vs UX vs latency vs cost) and own outcomes.
• Deep commitment to AI safety and the belief that safe interventions enable advanced model deployment.

Required skills

• Machine Learning systems
• Cloud platforms (multi‑provider deployments)