Research Program Manager, Safety Measurement

OpenAI•San Francisco, CA

7d•$239,000 - $328,000•Onsite

About The Position

The Safety Measurement Research Program Manager drives OpenAI’s strategy for safety measurement and evaluations, partnering with safety research, data science, and engineering to advance how harm and safeguard efficacy are measured across research and production. The role works with researchers, data scientists, policy owners, and product teams to identify important problems and shape the data, dashboards, tools, and workflows used to address them. It combines measurement expertise, product judgment, and cross-functional execution, including engagement with evaluation results and data workflows, prioritization tradeoffs, and delivery from problem definition through rollout and iteration. The position is based in San Francisco, California. Relocation assistance is available.

Requirements

At least 6 years of experience in research program management, technical program management, product management, or related roles.
Expertise in AI safety, trust and safety, integrity, or related domains.
A strong foundation in data science, statistics, economics, or a related field.
Hands-on experience with measurement or evaluations.
Ability to drive complex cross-functional programs independently from problem definition through rollout, building alignment and unblocking decisions in ambiguous situations.
Sound judgment about technical tradeoffs and the strengths and limitations of metrics and evaluations.
Ability to build partnerships across research, data science, engineering, policy, and product and communicate complex findings clearly to technical teams and leadership.
Curiosity about AI safety and its societal impacts, ability to learn new domains quickly, and adaptability as research and priorities evolve.

Responsibilities

Define strategic priorities and product roadmaps to improve safety measurement and evaluation approaches and scale the measurement platform to new use cases, products, and cross-functional team needs.
Partner with research, data science, engineering, policy, and other stakeholders to develop a vision for understanding safety outcomes and prevalence across products and translate it into concrete programs of work.
Integrate cutting-edge AI safety research into safety measurement and evaluation platforms, coordinating technical tradeoffs and dependencies and resolving blockers with researchers, data scientists, and engineers.
Develop and continuously refine clear, actionable success criteria for safety measurement and evaluations.
Represent quantitative progress on safety to senior leadership.