Research Engineer, Preparedness

Job Description

Job Title: Research Engineer, Preparedness. Location: San Francisco, CA

Responsibilities:

• Identify and explore emerging AI safety risks, with a focus on catastrophic misuse potential.
• Design and implement robust evaluation frameworks to assess risks in frontier AI models.
• Develop scalable tools and systems to support continuous AI safety assessments and evaluations;
• Lead projects end-to-end, from conception to deployment, contributing to AGI safety infrastructure;
• Collaborate across teams to refine risk management practices and establish safety evaluation best practices.
• Participate in internal red-teaming exercises and model observability initiatives.

Qualifications:

• Strong foundation in ML research engineering or development of ML-based applications.
• Experience with frontier models, safety evaluation tools, or red-teaming methodologies;
• Ability to work in a fast-paced, research-driven environment and manage cross-functional projects;
• Familiarity with societal and ethical considerations around AI deployment is a plus.
• Excellent problem-solving and communication skills.

Preferred:

• Red-teaming systems or adversarial testing.
• Working knowledge of AGI preparedness concepts or technical safety protocols.
• Experience with large language models and empirical model evaluations.

Benefits and Compensation:

• Competitive salary: $200,000 – $370,000
• Equity options;
• Work at the forefront of AI safety with OpenAI’s multidisciplinary team
• Inclusive and diverse work environment;
• Reasonable accommodations available for applicants with disabilities