PhD Researcher, Agentic 3D Modeling in Immersive Environments

Job Requisition ID # 26WD101374 Position Overview As a Research intern at Autodesk Research, you will explore and apply novel technologies and techniques to empower our customers in designing and building a better, safer, and more sustainable world.

We are a team of researchers, engineers, and industry domain experts working on projects that range from learning-based design systems, computer vision, graphics, robotics, human-computer interaction, sustainability, simulation, manufacturing, architectural design and construction.

Working with a multidisciplinary research team across AI, spatial computing, AEC, and human perception, you’ll develop and evaluate methods that interconnect different representations of the built environment within immersive, editable world models.

Possible research directions may include exploring editing non-traditional 3d geometries such as gaussian splats, building multi-layered spatial representations of architectural spaces (simulations, physics, rendering, reality capture), and controlling 3d modeling and design environments and processes with agentic workflows.

You’ll build research prototypes and explore how this work could inform future Autodesk design workflows.

Responsibilities

Research multimodal AI approaches for the generation and editing of novel 3d representations of built environments. Develop methods using foundation models, AI agents, and/or world models. Work with multimodal datasets, including visual, spatial, behavioral, and physics simulations data.

Design experiments and benchmarks to evaluate AI predictions against human responses. Build and evaluate research prototypes for real and simulated environments. Document and communicate findings through papers, presentations, and demonstrations.

Minimum Qualifications

Currently pursuing a PhD in Computer Science, Machine Learning, Computer Graphics, Computational Design, or a related field - with a completion date in 2028 or later. Research experience in AI/ML, computer vi