AI Research Scientist in Post-Training Alignment (Calgary)

Elevate your career at Autodesk as a Principal AI Research Scientist specializing in post-training alignment. This role emphasizes reinforcement learning and innovative algorithm development within a agile environment. At Autodesk, you will influence model development through post-training techniques like RLHF and RLAIF. Engage in designing experiments that shape model reliability, analyze behavior, and establish evaluation frameworks for long-horizon reasoning. Contributing to peer-reviewed publications and leading rigorous analysis efforts are key aspects of this senior-level research role. Key Responsibilities: - Develop algorithms for model reliability and alignment - Design experiments to enhance model behavior and robustness - Collaborate on scalable post-training workflows with infrastructure teams - Analyze model behavior and interpretability rigorously - Communicate technical trade-offs to leadership Requirements: - Expertise in reinforcement learning for foundation models - Proven leadership in technical research teams - PhD or extensive industry research experience - Strong publication record in leading AI venues - Familiarity with large-scale training infrastructure Leverage your deep expertise to advance AI research at Autodesk. Apply on Kit Job: kitjob.ca/job/2pjou1