A new paradigm for engineering design.

Braid believes that engineering design is the bottleneck of physical innovation — and that this bottleneck is the most consequential constraint on what humanity can build.

Every new engine, every lighter airframe, every more efficient heat exchanger begins as a design problem. Today, solving that problem takes far longer than it should. Not because engineers lack skill, but because the tools they depend on were designed for a different era. CAD knows nothing about physics. Physics knows nothing about manufacturing. The engineer is the only integration layer, stitching it all together through experience, intuition, and a great deal of manual effort.

The cost of this is not just slower design cycles. It is the long list of products that were never built, and the better versions of products we already have, because nobody had time to find them.

Imagine with us: aircraft structures whose geometry no human team would have considered drawing, but that are demonstrably lighter and more efficient. Heat exchangers redesigned to halve the energy they waste. Materials and manufacturing processes that can be matched to a problem in hours, not quarters. Engineering teams freed from the iterative grind, working at the level of intent and judgment — where their creativity actually matters.

When we founded Braid in 2020, the prevailing belief in AI was that scaling end-to-end machine learning on ever-larger datasets would solve nearly every problem. We took a different path. We believed that for AI to solve real engineering problems, where physical accuracy, safety, and manufacturability are non-negotiable, it would need to reason from the laws of physics and the constraints of manufacturing, not just learn patterns from data. The years since have validated that thesis: the AI systems producing reliable results in mathematical proof, and scientific discovery are the ones that combine learning with formal reasoning. We are applying the same paradigm to engineering design.

Our team is a tight braid of experts in the fields of physics, applied mathematics, computational geometry, large-scale simulation, and machine learning — the disciplines this problem demands. AURA, our system, is the first expression of what this combination makes possible.

This is what Braid was founded to build.

— The Braid founders

The technology behind AURA  →