ON-DEMAND WEBINAR
Addressing the Data Bottleneck for Foundational Physics Models
About this Webinar:
AI is transforming engineering workflows, with early implementations already reshaping how leading companies approach the design process.
In this webinar, we talk with Mohamed Elrefaie, a PhD student in the Schwarzman College of Computing and the Mechanical Engineering Department at MIT. Mohamed is an expert in the development of datasets for physics-based machine learning models. He is the creator of the DrivaerNet and DrivaerNet++ datasets–widely recognized as leading examples of open-source datasets for physics-based machine learning research. More recently, he has been focused on agentic AI and its impact on the automotive product development process. Mohamed has presented his innovative work at numerous prestigious organizations including Nvidia, Toyota Research Institute, MIT, Volkswagen Group, and Autodesk.
Join us for a virtual fireside chat exploring how machine learning is transforming engineering design, the vital role of high-quality datasets, and the implications of Physics AI for the future of automotive design.
What You'll Learn:
- How Physics AI and agentic workflows are converging to unlock faster, more intelligent design iterations
- Why the future of engineering design depends on diverse, high-quality datasets—and what defines such datasets
- How DrivaerNet++ was built, what it enables, and why it’s considered a benchmark for automotive ML research
- Mohamed’s perspective on Luminary’s SHIFT dataset and how it complements existing public datasets
- The emerging role of multi-modal data (e.g., images, meshes, CAD) in foundational model training for design and simulation tasks
Who Should Attend:
- Engineering and R&D leaders looking to leverage AI to reduce design cycle time and cost
- ML researchers and practitioners working on physical simulations or model training
- Simulation and CFD engineers interested in leveraging ML surrogate models for design exploration and optimization
- Product teams and technical executives evaluating the strategic value of foundational models in design
- Anyone curious about how agentic AI and Physics AI are reshaping engineering design
WATCH ON DEMAND
Meet the speakers
Mohamed Elrefaie, PhD Student, MIT
PhD Student
Massachusetts Institute of Technology
Mohamed Elrefaie is a recent graduate of the Technical University of Munich (TUM) with a bachelor’s in Mechanical Engineering and a master’s in Aerospace. He spent a year as a graduate research assistant at MIT's DeCoDE Lab. His research integrates deep learning with computational and experimental fluid dynamics to advance aerodynamic design. Currently, he is working on developing foundation physics models—AI models that can understand the physical world and apply this understanding to engineering design and simulations.
Mike Emory, PhD
Luminary Cloud
Product Manager
Mike Emory is the first product manager at Luminary Cloud, where he leads the SHIFT suite of physics-AI models. He holds a Ph.D. in Mechanical Engineering from Stanford University, where his research at the Center for Turbulence Research focused on uncertainty quantification for high-speed turbulent flows. Prior to Luminary, he spent eight years at Cascade Technologies (now Cadence) advancing large-eddy simulation tools for industrial applications in aerodynamics, combustion, and heat transfer. His current work centers on building simulation datasets to power machine learning models for fast, accurate physics prediction.
Joe Warner
Luminary Cloud
Product Marketing Manager
Joseph Warner is a product marketer at Luminary Cloud, where he helps shape the go-to-market strategy for physics-based machine learning. He holds both a B.S. and M.S. in Mechanical Engineering from the University of Tennessee, where his graduate work at Oak Ridge National Laboratory focused on novel technologies for net-zero energy buildings. Before joining Luminary, he began his career at Siemens as a Solutions Consultant, specializing in CFD for thermal applications. Today, he works closely with customers to showcase how advanced simulation and AI are accelerating innovation in product development.