Principal AI Research Scientist Post-Training Alignment
Core
Leading post-training alignment research for foundation models using reinforcement learning, preference optimization, and agentic systems to improve reliability and controllability within Autodesk's architecture, engineering, and manufacturing domains.
Role type
Principal AI Research Scientist (Post-Training Alignment)
Builds
Scalable post-training workflows, novel alignment algorithms, and evaluation frameworks for long-horizon reasoning and agentic behavior.
Domain
AI Research / Machine Learning / Reinforcement Learning / Post-Training Alignment
Deliverable
production ML models
Required skills
Reinforcement learning for foundation models, post-training methods (RLHF, RLAIF, DPO, PPO), leading technical research teams, model behavior intuition, evaluation system design, technical communication, production AI system deployment, large-scale training infrastructure knowledge
Preferred skills
None stated
Technologies
None stated
Responsibilities
Develop novel algorithms for model reliability and alignment, design and run experiments on model behavior and reasoning quality, partner with infrastructure teams to build scalable post-training workflows, contribute to publications and patents, design evaluation frameworks for agentic behavior and safety, lead model analysis and interpretability efforts, drive human-in-the-loop evaluation, establish model readiness criteria, communicate technical risks to leadership
Seniority
Principal, strategy & mentorship
