Normal Computing develops software and hardware designed to reinvent computing for artificial intelligence by aligning hardware design with the intrinsic properties of physical systems. The company's approach centres on two main offerings: Normal EDA, an AI-accelerated co-design platform for silicon engineering teams, and Normal ASICs, physics-based application-specific integrated circuits built using thermodynamic computing that the company claims deliver up to 1,000 times more intelligence per unit of energy.
The company was founded by former members of Google Brain and Google X, and its team draws from Google Brain, Google X, Meta, Palantir, IBM Research, UC Berkeley, NVIDIA, and Intel. Normal Computing maintains offices in New York City, San Francisco, London, and Copenhagen, and works in partnership with major semiconductor design and manufacturing institutions.
Normal Computing operates at the intersection of hardware design, semiconductor manufacturing, and AI infrastructure. Its technical focus spans AI-accelerated electronic design automation, thermodynamic computing, quantum AI, physics-based computing, and ASIC design. The company has developed machine learning frameworks for probabilistic and quantum AI applications.