NomadDesk

AI Research Manager/Scientist, Model Alignment

Job Requisition ID #

26WD94883

Position Overview

As an AI Scientist Manager at Autodesk Research, you will be doing fundamental and applied research that will help our customers imagine, design, and make a better world. We are seeking an AI Scientist Manager to lead our post-training and model alignment efforts. This role sits at the critical intersection of advanced AI research, people leadership, and model readiness. You will both manage and grow a team of AI scientists and personally contribute as a hands-on researcher, owning the transformation of foundation models into reliable, aligned, and production-ready systems. This is not a purely managerial role. You will remain deeply technical while setting direction, making trade-offs, and taking accountability for model behavior at release.

Autodesk's AI Lab is active in the wider research community, targeting publications at CVPR, NeurIPS, ICML, ICLR, SIGGRAPH, and other top-tier conferences. We collaborate with top academic & industry labs, combining the best of an academic environment with product-guided research. We are a global team, located in London, San Francisco, Toronto, and remotely in the US, Canada, and Europe.

This role will report to Director of AI Research in the AI Lab.

Key Responsibilities

Technical Leadership & Hands-On Research

• Lead and contribute directly to post-training pipelines, including:
  
  
  • Instruction tuning and multi-task fine-tuning
  • Preference optimization (RLHF, RLAIF, DPO, PPO, and related methods)
  • Domain-specific post-training and specialization for the AECO, Manufacturing, and Media & Entertainment industries
• Design and run experiments that shape model behavior, robustness, and reliability
• Decide what problems are best addressed through post-training vs pre-training vs product-level mitigation
• Partner with infrastructure teams to ensure efficient, reproducible, and scalable post-training workflows

• Design and maintain evaluation frameworks that measure:
  
  
  • Long-horizon reasoning and planning
  • Tool-use and agentic behavior
  • Safety, robustness, and alignment
  • Regression and behavioral drift across releases
• Lead human-in-the-loop evaluation, ensuring annotation quality, consistency, and bias awareness
• Provide clear go / no-go recommendations for model releases, including explicit articulation of known risks and trade-offs

• Manage, mentor, and grow a team of AI scientists working on post-training and alignment
• Set clear technical direction while empowering researchers to own end-to-end projects
• Hire and develop scientists with strengths across ML, RL, evaluation, and human-centered AI
• Foster a culture of:
  
  
  • Rigorous experimentation and ablation
  • Reproducibility and scientific integrity
  • Thoughtful risk-taking and humility about model behavior
• Provide regular feedback, career coaching, and performance management

• Act as a key interface between:
  
  
  • Pre-training research
  • Infrastructure and compute teams
  • Model Delivery team
  • Safety, policy, and legal stakeholders
• Translate complex research trade-offs into clear, decision-ready guidance for leadership
• Influence the broader AI roadmap by identifying post-training opportunities that unlock product impact

• PhD or equivalent industry experience in Machine Learning, AI, or a related field
• Proven experience as a people manager of technical research or ML teams
• Strong hands-on expertise in:
  
  
  • Large language models or foundation models
  • Fine-tuning and post-training methods (e.g., RLHF, DPO, instruction tuning)
  • Experimental design and evaluation
• Ability to move fluidly between research depth and organizational leadership
• Strong communication skills, with the ability to explain complex trade-offs to technical and non-technical audiences

• Experience operating in an AI research lab or frontier model organization
• Background in human-in-the-loop systems, preference learning, or alignment research
• Experience shipping or supporting production AI systems
• Familiarity with large-scale training infrastructure and compute cost trade-offs
• Experience in Architecture, Civil or Mechanical Engineering, Construction, Manufacturing, Media & Entertainment or other Autodesk domains

• Post-trained models demonstrate measurable improvements in reliability, alignment, and usefulness
• Evaluation metrics are trusted and adopted across teams
• The team consistently delivers high-quality research and practical impact
• Leadership relies on your judgment for model readiness and risk assessment
• Team members grow into strong, independent researchers and leaders