Research Scientist (Safety Post Training)

Join Scale AI as a Research Scientist focused on Safety Post-Training. In this role, you will develop and apply post-training methods and interpretability techniques to enhance the safety and understanding of frontier AI systems. You will design and run post-training pipelines, develop interpretability-informed evaluations, and collaborate with policymakers, engineers, and researchers to translate findings into actionable safety standards and best practices.

• Design and run post-training pipelines to study how training choices affect model safety, robustness, and alignment properties.
• Develop interpretability-informed evaluations that reveal how and why models produce unsafe, deceptive, or otherwise undesirable behaviors.
• Collaborate with policymakers, engineers, and other researchers to translate post-training and interpretability findings into actionable safety standards.
• At least three years of experience addressing sophisticated ML problems, whether in a research setting or in product development
• Commitment to our mission of promoting safe, secure, and trustworthy AI deployments in the industry as frontier AI capabilities continue to advance
• A track record of published research in machine learning, particularly in generative AI
• Experience with post-training and RL techniques such as RLHF, DPO, GRPO, and similar approaches
• Strong written and verbal communication skills to operate in a cross-functional team
• Experience studying failure modes introduced or masked by post-training, such as reward hacking, sycophancy, or alignment faking
• Experience with mechanistic interpretability, probing, or other techniques for understanding model internals
• Familiarity with red-teaming or adversarial evaluation of post-trained models
• We will not ask any LeetCode-style questions. If you’re excited about advancing AI safety and contributing to our mission, we encourage you to apply, even if your experience doesn’t perfectly align with every requirement