Principal AI Research Scientist Post-Training · Alignment · Reinforcement Learning Autodesk AI Lab: London · San Francisco · Toronto · Remote (US/CA/EU
Core
Leading post-training and alignment research for foundation models using reinforcement learning, focusing on reliability, controllability, and long-horizon reasoning within Autodesk's engineering and design domains.
Role type
Principal AI Research Scientist (Post-Training & Alignment)
Builds
Production-ready aligned foundation models and agentic systems for architecture, engineering, and manufacturing workflows.
Domain
AI Research / Reinforcement Learning / Computer-Aided Design (CAD)
Deliverable
production ML models
Required skills
Reinforcement learning for foundation models, Post-training methods (RLHF, RLAIF, DPO, PPO), Model alignment and safety, Long-horizon reasoning, Experimental design, Technical leadership and mentoring, Evaluation framework design, Model interpretability, Human-in-the-loop evaluation, Production AI system deployment
Preferred skills
Experience at frontier model labs, Strong publication record at top ML venues, Familiarity with large-scale training infrastructure
Technologies
RLHF, RLAIF, DPO, PPO, CAD kernels, Physics simulation engines
Responsibilities
Develop novel algorithms to improve model reliability and alignment, Design and run experiments shaping model behavior, Partner with infrastructure teams to build scalable post-training workflows, Lead rigorous model analysis and interpretability efforts, Establish model readiness criteria for releases, Communicate technical risks to leadership
Seniority
Principal, strategy & mentorship
