Meet Synopsys.ai Copilot, Industry's First GenAI Capability for Chip Design

Freeing Up Time for Product Differentiation

Semiconductors have become increasingly complex in the face of compute-intensive applications such as AI, high-performance computing, and autonomous automotive. Vertically stacked multi-die systems, devices featuring billions of transistors, and angstrom-scale structures are making waves across the industry. At the same time, just when their expertise is in such great demand, an engineering talent shortage is threatening to stall the innovation that has led us where we are now.

AI provides a solution, taking on repetitive tasks to free engineers to focus on product differentiation. Tapping into transformer architecture in large language models (LLMs), GenAI in particular can accomplish in seconds what would take someone hours or even days.

Synopsys.ai Copilot brings GenAI to the design engineering masses. Regardless of their level of experience, every designer can take their productivity up across every phase in the chip design cycle. It’s like having an expert guide at your fingertips. Initially, Synopsys.ai Copilot will serve as a knowledge query system, pulling from Synopsys resources such as product user manuals, application notes, videos, and any documentation found on the Synopsys SolvNetPlus support community. Queries can range from simple questions (“Which option should I use?”) to more complex and specific questions that might entail follow-up questions by Copilot (“How would I shield those nets in my design ?”). Since the Copilot will work within the confines of a particular organization, it will learn to be able to provide contextual responses. For example, say an engineer is seeking a specific capability in Synopsys Fusion Compiler™. In forming the response, Synopsys.ai Copilot would observe what the engineer is working on to provide a more relevant answer. Traditionally, the engineer would research answers on software manuals or SolvNetPlus, seek advice from an experienced co-worker, or even call on an applications engineer—all of which takes much more time and effort.

Over time, as it is enriched and continues to learn more from your unique workflow and methodologies—in a secure, closed-loop manner—the Synopsys.ai Copilot will be able to provide prescriptive guidance and recommendations and also create workflow scripts. For example, it might help with debug, identify the worst timing path and provide a suggestion to fix it, or offer a script to apply to a particular cell to optimize for power. What’s more, these results could be served up in just minutes.

How would an engineer know whether the answers are accurate? That’s where the human comes into play, providing the checks and balances. After all, the intention of Synopsys.ai Copilot is to help the engineer get the job done faster.