Fork the consciousness, or download the project and create your own.

Patrick Coles

Chief Scientist, Normal Computing

Builder The claim is about what the hardware does, and the philosophy is left open

Patrick Coles is the chief scientist of Normal Computing, the company that taped out the CN101 thermodynamic chip in June 2025. He is the senior author of Thermodynamic linear algebra, published in npj Unconventional Computing in November 2024 with Maxwell Aifer, Kaelan Donatella, Max Hunter Gordon, Samuel Duffield, Thomas Ahle, Daniel Simpson and Gavin Crooks. The paper shows that a network of coupled harmonic oscillators, left to reach thermal equilibrium, solves linear systems, inverts matrices, computes determinants and solves Lyapunov equations, with an asymptotic speedup over digital methods that grows linearly with the matrix dimension.

He is also the senior author of the first hardware demonstration, a stochastic processing unit of eight coupled electrical oscillators on a printed circuit board that inverted an 8 × 8 matrix and sampled Gaussian distributions. At the CN101 announcement he set out the roadmap from that chip to medium-scale generative AI on CN201 and large-scale generative AI on later chips. The work is reviewed in Normal Computing CN101 and thermodynamic linear algebra.

He is covered here under the Builder position, for the algorithms and the hardware. No view on consciousness is attributed to him.

Known for. Thermodynamic linear algebra (npj Unconventional Computing, 2024), the first thermodynamic matrix inversion on a circuit board (2023), and the scientific roadmap of Normal Computing's CN101 chip

Coverage on this site

Related researchers