Technical Program Manager, AI Safety & Safeguards

Responsibilities

• Partner with product, engineering, research, and design to shape technical roadmaps, platform direction, and safeguards priorities across ChatGPT, API, enterprise, and cloud environments.
• Translate product and safety objectives into execution plans balancing delivery, platform evolution, customer needs, and responsible deployment.
• Lead cross-functional programs from problem definition and technical design through implementation, launch, iteration, and operational follow-through.
• Own milestones, dependencies, risk mitigation, decisions, and accountability across engineering, infrastructure, integrity, trust and safety, legal, policy, operations, and go-to-market teams.
• Partner with engineers on system architecture, APIs, cloud deployment patterns, platform integrations, operational failure modes, and technical tradeoffs.
• Drive safeguards including model evaluations, post-training mitigations, classifiers, abuse detection, monitoring, enforcement workflows, human review, and agent-assisted review.
• Coordinate readiness for sensitive or high-impact deployments, including safety coverage, escalation paths, operational controls, and partner alignment.
• Identify emerging misuse and severe-harm risks and define prevention, detection, investigation, enforcement, and improvement mechanisms.
• Establish metrics for program delivery, deployment readiness, safety effectiveness, operational quality, and user or developer experience.
• Communicate program health, risks, recommendations, and tradeoffs to engineering teams, cross-functional partners, and executives.
• Improve delivery and safety outcomes through operating mechanisms, scalable tooling, and data- or AI-assisted workflows.

Requirements

• Strong track record delivering high-impact technical programs across product, platform, API, infrastructure, machine learning, or safety domains.
• Strong technical judgment across distributed systems, cloud deployments, public or internal APIs, developer platforms, production machine learning, or scalable product architecture.
• Ability to engage credibly with engineers on system design, implementation tradeoffs, reliability, deployment constraints, and operational risks.
• Experience delivering platforms or developer-facing capabilities and understanding how technical decisions affect customer and developer experience.
• Understanding of safety, integrity, or abuse-prevention systems, including evaluations, guardrails, classifiers, monitoring, enforcement, investigations, and incident response.
• Strong program management discipline, including coordinating workstreams, resolving dependencies, and turning ambiguity into accountable execution.
• Product judgment balancing user and customer needs, model usefulness, safety risk, technical complexity, and business priorities.
• Clear and empathetic communication with engineering teams, executives, policy and legal partners, and external stakeholders.
• Comfort owning high-stakes programs requiring executive presence, cross-organizational influence, and decisions under uncertainty.
• Relevant experience with sensitive deployments, major cloud providers, or serious-harm domains such as violent misuse, child safety, or cybersecurity.
• Hands-on, curious approach to responsible deployment of increasingly capable AI systems.