Accelerating Innovation: AI-Driven Advances in Sigrity, Clarity, and Optimality.

Qin Liu is a Software Engineering Director in Sigrity R&D US Group at Cadence Design Systems, mainly focused on Clarity 3D full-wave electromagnetic software development. Dr. Liu has over 14 years’ experience in developing advanced electromagnetic analysis and simulation methods. She received the B.Eng. degree in electronic engineering from the University of Science and Technology of China, Hefei, China, in 2011, and the Ph.D. degree in electrical and electronic engineering from The University of Hong Kong, Hong Kong, in 2015. From 2015 to 2018, she is a Post-Doctoral Fellow with The University of Hong Kong and a Postdoctoral Research Associate with the University of Illinois at Urbana Champaign. Since 2018, she has been working for Sigrity/Clarity R&D Group at Cadence Design System, San Jose, USA.

With the electronic systems being more complex, efficiency and accuracy requirement on design and simulation is becoming more and more demanding. At Cadence, we’re meeting this challenge head-on by bringing AI and automation to every stage of design. Sigrity-APX, as an intelligent advanced IC Package Extractor, embeds machine learning model in many typical but challenging structures, such as vias, degassing hole planes, and traces with degassing hole planes. Clarity-PI is also launched as an AI accelerated Clarity extraction tool with faster, smarter and more stable solution, benefitting IC, interposer, and packaging extraction applications. Also, with new statistical functions and AI surrogate models, our advanced optimization engine allows user to explore the design space with higher efficiency. Furthermore, major break through is made in using generative AI beyond traditional optimization, we can now automate the entire workflow inside the design platform by creating new design from scratch, identifying critical regions, auto-cutting and extracting, and running targeted optimizations. AI algorithm, learning from Cadence’s massive simulation datasets, predicts electromagnetic effects with high fidelity. This means fewer manual tweaks and more reliable results, accelerating the path to signoff.