Building Principled and Generalizable Intelligence for Engineering Design
- Sponsor
- Mechanical Science and Engineering
- Speaker
- Mr. Lyle Regenwetter, Mechanical Engineering, Massachusetts Institute of Technology (MIT)
- Contact
- Amy Rumsey
- rumsey@illinois.edu
- Phone
- 217-300-4310
- Originating Calendar
- MechSE Seminars
Abstract
Generative AI has reshaped how we write, code, and illustrate—but not how we design. Current models produce statistically plausible outputs, yet engineering design requires precise constraint satisfaction, human-centered integration, and generality across domains. First, I examine the tension between probabilistic models and the reliability demands of high-stakes engineering. I introduce a counterexample-driven training paradigm that yields order-of-magnitude gains in constraint satisfaction and data efficiency. Second, I illustrate how AI's disconnect from design practice limits its adoption. I present human-centered benchmarks and methods that forego end-to-end design in favor of small, surgical design modifications. Third, I showcase synthetic-data pipelines that allow models to acquire general engineering knowledge, enabling a single model to solve disparate design problems. Together, these advances demonstrate AI systems that satisfy constraints, operate within design workflows, and generalize across domains—extending generative AI beyond statistical plausibility toward engineering-grade intelligence.
About the Speaker
Lyle Regenwetter is a Ph.D. Candidate in Mechanical Engineering at MIT. He received his S.M. in Mechanical Engineering from MIT in 2022, as well as a B.S. in Electrical Engineering and a B.S. in Mechanical Engineering with highest honors from the University of Illinois at Urbana-Champaign in 2020. Lyle has received numerous academic distinctions, such as the Morningside Academy for Design Fellowship at MIT, as well as the Chancellor's scholarship, Association for Facilities Engineering Award, and Kenneth J. Trigger award at UIUC.
Host: Professor Bill King