Calibre Directions in Artificial Intelligence

Engineering & simulationCalibre Vision AIshrunk verification time by 50%

AI organized millions of DRC errors into tens of signals, allowing designers to assign groups of errors to specific owners and reducing verification cycle time.

“So the AI component of vision AI is to organize these 1000000s of errors into common areas or potentially even the same common root causes. And we're able to organize those 1000000s of errors into 10s of what we call signals. So that in and of itself is a huge productivity gain … the net result of this story is they estimated that they shrunk that verification time by 50…”

Engineering & simulationCalibre RVE Check Assist

An AI agent sourced root cause information from foundry manuals and internal best practices to help designers understand and fix DRC errors, serving as on-the-job training for junior staff.

“So the Czech assist capability goes back to either a foundry, uh, design room manuals. Or best practices within the given company about what is, what drove that to be an error. And so it's essentially advanced documentation and advanced isolation of that root cause to help the designer determine, okay, here's here's the nature of the error.”

Engineering & simulationSiemens EDA AI System

AI techniques identified the most efficient use of hardware resources (on-premise, cloud, or hybrid) to set up and run verification tasks, optimizing job execution speed.

“we provide. a number of techniques to identify what is the most efficient use of your hardware resources. To get a particular verification task done. So that's where we start, is helping you to set up a particular caliber run and leverage what hardware resources you have available to you, to get this job done as quickly and efficiently as possible.”