An Engineer’s Approach to Integrating Machine Learning in Generative Design Tools

CDFAM Computational Design Symposium — Barcelona 2026

Thomas Rees · ToffeeX

Machine learning and Artificial Intelligence have enormous potential as tools for generative design in engineering. However, most industry efforts remain stuck in research prototypes, brittle bespoke models, or disconnected add-ons that rarely survive real engineering workflows. In this talk, we will present an engineer’s approach to integrating machine learning directly into production-ready generative design tools, drawing on our experience building the fastest physics-driven thermo-fluid optimization platform on the market. Rather than replacing physics with opaque black boxes, our methodology uses ML only where it strengthens engineering outcomes.

I will show how ToffeeX’s ML developments accelerate design exploration and automation while preserving full control of engineering intent, seamlessly extending our existing topology optimization engine which is already used daily in real production environments. This talk highlights why our approach, built on smart algorithmic design rather than brute-force model training, achieves the reliability, manufacturability, and speed required for real-world engineering.

Links

Talk page with full transcript

Watch the talk on YouTube

Cite this talk

Search the full text of every recorded CDFAM presentation

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.designforam.com