Webinar – IP Design Considerations for Real-Time Edge AI Systems

It is well-known that semiconductor growth is driven by AI. That simple statement breaks down into many complex use cases, each with its own requirements and challenges. A webinar will be presented by Synopsys on October 23 that focuses on the specific requirements for one of the most popular use cases – AI at the edge. The speaker is very knowledgeable on the topic and will treat the audience to a comprehensive view of the many requirements to be considered for successful deployment of AI at the edge. I highly recommend you register for this important event. A link is coming but first let’s look a little closer at this webinar on IP design considerations for real-time edge AI systems.

The Presenter

The value of any webinar is heavily influenced by the presenter. In this case, it’s Hezi Saar, executive director of product line management for mobile, automotive, and consumer IP for the Synopsys Solutions Group. Hezi brings more than 20 years of experience in the semiconductor and embedded systems industries.  He has been with Synopsys for almost 17 years. He has also been the Chair of the Board of Directors for the MIPI Alliance for over nine years. Before Synopsys, Hezi was involved in product marketing, product management and design at Actel, ISD/Winbond and RAD Data Communications.

Hezi is quite good at explaining complex topics in an easy-to-understand way. The 25-minite webinar will be followed by a Q&A session with questions from the audience. I’m sure Hezi will do a great job with those questions as well.

Some Webinar Topics

Hezi presents a broad overview of IP architecture and integration methodologies that support real-time AI workloads at the edge. Here are some other topics he discusses:

A useful historical perspective of how AI has driven semiconductor growth is presented. The trends associated with AI models are discussed – the focus is to provide more capacity with less resources. How the quality of small models for edge AI has increased is also reviewed. The motivations for moving from the cloud to the edge is another interesting topic. Power efficiency is critical here. Hezi presents data that shows power consumption can be up to 200X more efficient on the device (edge) vs. the cloud.

There are many drivers and many benefits associated with moving from the cloud to the edge. He points out that this is what’s driving the next innovation cycle as summarized by the diagram below.

Hezi then presents a very useful and informative overview of a broad range of smart and connected devices at the edge. He discusses the unique requirements for cost, performance, area and power for these cases.

Market data is also presented, showing edge Al device shipments for smartphones dominating, with smart speakers showing growth as well. There is a lot of very useful discussion around new opportunities and how to address consumer demands for cost-effective products. Considerations for model choices are discussed, along with an overview of how AI companion chips can help.

Hezi also explores the impact of multi-die approaches. This technology will help in some cases more than others.

To Learn More

I have touched on only a subset of the topics Hezi covers in this very informative webinar. If AI at the edge is in your plans, this webinar will provide substantial and valuable information. I highly recommend investing the time to attend. It will take less than an hour.

The webinar will be held Thursday, October 23, 2025, from 10:00 AM – 11:00 AM Pacific Daylight Time. You can reserve your spot at the event here. And that’s a summary of a webinar on IP design considerations for real-time edge AI systems.

Also Read:

Synopsys and TSMC Unite to Power the Future of AI and Multi-Die Innovation

AI Everywhere in the Chip Lifecycle: Synopsys at AI Infra Summit 2025

Synopsys Collaborates with TSMC to Enable Advanced 2D and 3D Design Solutions