Agent Safety Researcher: Training & Evaluation

CHEManager International; San Francisco (CA); Hybrid; USD 190,000 - 240,000; Full time

Job summary

OpenAI is seeking exceptionally capable researchers and engineers to advance frontier models while ensuring safety and alignment with user intent. The role emphasizes independent execution, ambiguity tolerance, and a strong foundation in modern frontier-model research.

Based in San Francisco with a hybrid work model, relocation support is available for the right candidate. You will collaborate across teams to deploy research-backed mitigations in large-scale training and agent systems.

Qualifications

• Strong background in research engineering, ML engineering, quantitative research, or applied model research.
• Ability to own ambiguous projects end to end.
• Excellent execution across experimentation, data, evaluation and infrastructure.

Responsibilities

• Train and evaluate frontier models to reduce harmful or misaligned agent actions and form hypotheses.
• Mine incidents and build scalable measurement, data-processing, and evaluation systems.
• Collaborate with post-training, capabilities, oversight, and pre-training teams to ship mitigations into large-scale training and agent systems.

Skills

Research engineering

ML engineering

Quantitative research

Applied model research

Tools

Data pipelines

Experimentation infra

Job description

OpenAI is seeking exceptionally capable researchers and engineers to advance frontier models while ensuring safety and alignment with user intent. The role emphasizes independent execution, ambiguity tolerance, and a strong foundation in modern frontier-model research.

Based in San Francisco with a hybrid work model, relocation support is available for the right candidate. You will collaborate across teams to deploy research-backed mitigations in large-scale training and agent systems.