About this role
Scale AI is seeking a Research Scientist to focus on AI safety post-training. You will develop and apply methods to make frontier AI systems safer and more understandable for researchers and policymakers. This role involves designing experiments, developing evaluations, and collaborating with cross-functional teams to translate findings into safety standards and best practices.
What we are looking for 6
- Develop and apply post-training methods for AI safety
- Design pipelines to study model safety, robustness, and alignment
- Develop interpretability-informed evaluations for undesirable behaviors
- Collaborate with policymakers, engineers, and researchers
- Translate findings into actionable safety standards and benchmarks
- Experience with post-training and RL techniques (e.g., RLHF, DPO)