PhysicsX Ltd., a startup using artificial intelligence to speed up hardware design projects, today announced that it has raised $300 million in funding.

Returning investor Temasek led the Series C round. It was joined by Nvidia Corp., chipmaking equipment supplier Applied Materials Inc., Siemens AG and others. The investment values PhysicsX at $2.4 billion.

Engineers design hardware products such as chips and car parts through an iterative process. They create an initial design, test it in a simulation and use the information gleaned during the evaluation to refine the blueprint. They then repeat the workflow many times until arriving at an optimal version of the design.

Turning an initial set of blueprints into a market-ready product can take years. London-based PhysicsX offers a platform that uses AI to speed up the process. According to the company, its software can perform tasks that usually take hours or days in a few seconds.

Under the hood, an engineering simulation is a sequence of partial differential equations, or PDEs. Such equations can model the physical conditions under which a system is expected to operate and the manner in which it will perform. PDEs take a significant amount of time to solve using traditional methods.

According to PhysicsX, the platform speeds up the task with algorithms dubbed neural operators. A neural operator is an AI model that skips some of the steps involved in solving a PDE. It does so by reusing the results of similar equations that were solved in the past instead of calculating everything from scratch.

“Almost every hard problem in the physical economy — better aircraft, better chips, better engines, better energy systems — comes down to how fast and how well engineers and machine operators can work through the underlying physics,” said PhysicsX co-founder and Chief Executive Officer Jacomo Corbo.

The company trains some of its AI models on internally-produced engineering datasets. PhysicsX’s car simulation algorithms, for example, use vehicle designs it generates using a tool dubbed the Data Factory. The software automatically creates variations of 260 baseline designs curated by the company’s staffers.

Some of the platform’s AI models help engineers optimize the shape of car parts to make them more aerodynamic. Others focus on multiphysics simulations, which are used to study how the physical phenomena that affect a system interact with one another. For example, a chipmaker might wish to evaluate the correlation between the strength of the current that runs through a chip and the amount of heat it generates.

PhysicsX offers its AI models alongside an array of other engineering tools. A feature called the Simulation Workbench helps engineers organize the data produced by simulations. Additionally, PhysicsX provides dozens of templates that ease the task of customizing its platform to a company’s engineering projects.

The company says its software is used by customers in more than a half-dozen verticals, including the semiconductor, industrial machinery and auto sectors. PhysicsX claims to have tripled its booked revenues in the past year thanks to strong demand. The company’s headcount doubled to 300 employees.

It will use the proceeds from its Series C round to continue its growth. It plans to expand its international presence and train more capable AI models.

Photo: Unsplash