AI platforms for composite materials are starting to do for materials what CAD did for mechanical design: let engineers explore thousands of variants before touching a lab. These systems combine physics-informed models, historical test data, and optimisation algorithms to propose layups, fibre orientations, and resin systems tuned to specific constraints like weight, stiffness, cost, or temperature. The real breakthrough is not only faster simulation but closed-loop learning, where new test results feed the model and refine future predictions, cutting iteration cycles dramatically. For SMEs in sectors like mobility or wind energy, partnering with such a platform is often smarter than building in-house: you get access to up-to-date models, compliance templates, and cloud compute without hiring a full data science team.
Background in materials science + AI: My work focuses on the intersection of computational materials design and applied machine learning. Over the past decade, I've worked with teams using AI to shorten the composite development cycle — not by predicting properties, but by identifying why a microstructure behaves the way it does. That distinction is where the real breakthroughs are happening. What I can contribute to the podcast: The most transformative shift I'm seeing is that AI platforms are no longer just screening candidate materials; they're generating structure-property explanations that scientists can actually use. When the algorithm shows how fiber orientation, matrix chemistry, and manufacturing tolerances interact, it changes the workflow entirely. You don't just get a "better composite." You get a blueprint for how to build it repeatably. That's the part of the story most people miss — AI isn't replacing materials scientists; it's giving them visibility into interactions that were previously invisible or too expensive to test. I'd be happy to dive into real case examples where this has accelerated development by months and dramatically reduced failed prototypes.