Staff / Principal Research Engineer, AI Safety Mitigations

What you'll do

• Translate risk assessments, policies, and evaluation findings into mitigation designs and testable requirements; implement safeguards across relevant systems.
• Run experiments to measure risk reduction, robustness, and impact on legitimate scientific use, and use adversarial findings and failures to improve interventions.
• Integrate safeguards with model-training, post-training, inference, and platform teams; investigate regressions, distribution shifts, and misuse patterns.

What they ask for

• Hands-on experience developing and validating safety interventions for frontier models in training, post-training, or large-scale inference, with contributions to model development or release cycles.
• Depth in at least one area: safety post-training, model-level classifiers, inference safeguards, or abuse monitoring and enforcement across a model-serving platform.
• Experience working with model-development or safeguards teams and using relevant system access to investigate failures and implement improvements.
• Strong research-engineering skills, experimental judgment, and ability to collaborate across research and engineering teams.

Nice to have

• Experience with scientific models, scientific software, or laboratory-connected systems.
• Safety work involving tool-using agents, multi-step workflows, or high-consequence misuse; or connecting expert evidence and policy to training data, evaluations, or operational controls.

Pay

$224,000 — $336,000 USD (base salary)

AI focus

Develops and evaluates safety mitigations for frontier and scientific AI, including model training, classifiers, monitoring, access controls, runtime interventions, and oversight-related research.