Skip to content
Advancing Bits to Master Atoms

Research at MatterSpace.
Two directions, each with its own product.

MatterSpace runs two separate research directions. One studies self-improving AI for science, research and engineering. The other studies energy-based search and optimization, for design spaces far too large to try one candidate at a time.

Two precision instruments side by side on a long graphite plinth: a satin-silver cylindrical housing with a cobalt-blue core seen through a window, and a satin-silver cubic lattice frame with cobalt-blue spheres at its nodes.Two precision instruments side by side on a long graphite plinth: a satin-silver cylindrical housing with a cobalt-blue core seen through a window, and a satin-silver cubic lattice frame with cobalt-blue spheres at its nodes.

Two research directions.

Each direction has its own questions, its own page and its own product.

  • Self-improving AI

    Neuro-Analytic AI

    Neuro-Analytic AI pairs the AI you already use with a complementary analytical AI that learns from validated outcomes and helps you decide what to try next.

    Product: MS Duplex

  • Energy-based search and optimization

    Adaptive Learning, Searching and Optimization

    How to search very large design spaces for candidates that meet hard requirements, and how to spend expensive computation where it is most likely to pay off.

    Product: MS Lattice

Where the research goes next.

What we delivered in 2026, and the one new research area we plan for 2027: chip design.

Talk to us about our research.

If your team works on a problem where better decisions or a better search would change the outcome, we would like to hear about it.