Staff / Principal Research Engineer, AI Safety Mitigations

Highlights

• Build safeguards that reduce societal risks from closed-loop scientific AI.
• Turn risk assessments and eval findings into testable mitigation designs.
• Integrate mitigations into training, release evaluation and large-scale serving.

What they offer

• Comprehensive benefits program tailored to region

Ideal candidate

A high-agency research engineer who has built frontier-model safety interventions and wants to apply that to science. You thrive on ambiguous problems, rigorous experimentation, and collaborating across science and engineering to ship reliable safeguards.

Requirements

• Hands-on experience developing and validating safety interventions for frontier models in training, post-training or large-scale inference systems
• Contributions to model development or release cycles
• Depth in safety post-training, model-level classifiers, inference-integrated safeguards, or abuse monitoring and enforcement
• Experience working with model-development or safeguards teams using training pipelines, model internals or inference infrastructure to investigate failures
• Strong research-engineering skills moving from ambiguous problems to working, maintainable implementations
• Sound experimental judgment assessing robustness and reasoning about competing objectives
• Ability to collaborate across research and engineering teams and own complex individual contributions