Research Scientist, Frontier Risk Evaluations
Core
Design and create evaluation measures, harnesses, and datasets to measure risks posed by frontier AI systems, including testing for dangerous capabilities like security vulnerability exploitation and CBRN uplift.
Role type
Research Scientist (AI Safety & Risk Evaluation)
Builds
Evaluation harnesses, datasets, and technical reports for policymakers and government agencies.
Domain
Artificial Intelligence Safety, Policy Research, and Risk Mitigation
Deliverable
production ML models | research | client delivery
Required skills
ML pipeline instrumentation, evaluation harness design, generative AI research, technical report writing, cross-functional collaboration
Preferred skills
Red-teaming, adversarial testing, AI safety policy frameworks (NIST AI RMF, EU AI Act)
Technologies
Generative AI, ML pipelines
Responsibilities
Design and build harnesses to test AI models for dangerous capabilities; Collaborate with government agencies to scope and design risk evaluations; Publish evaluation methodologies and technical reports for policymakers.
Seniority
Mid-Senior, hands-on IC
