Future-Proofing AI Chips for Edge & Cloud

  July 15th 2025                                                                      Duration: 1 hour

As AI workloads increasingly dominate numerous cloud and edge applications, building a scalable, efficient, and future-ready AI chips is no longer optional—it’s essential. This webinar explores the latest AI trends, highlighting the growing demand for intelligent processing closer to the data source and the critical role of NPUs in enabling this shift.

We’ll dive deep into what it means to future-proof AI chips, focusing on architectural flexibility, power efficiency, and the ability to scale across diverse applications—from smart cameras and autonomous systems, to cloud computing. At the heart of this discussion is the Ceva-NeuPro-M scalable NPU IP architecture, a cutting-edge solution designed to meet the evolving needs of AI in cloud and at the edge. Learn how the NeuPro-M modular design, advanced SDK support, and multi-core scalability empower developers to stay ahead of the curve in a rapidly changing AI landscape.

What you will learn in this session:

• Key trends shaping AI today – from generative models to agentic systems, and their impact on future product and system design
• How modern NPU architecture delivers high performance with power efficiency, optimizing compute, memory, and precision per target device constraints
• How NPU modular design ensures scalability and adaptability for future AI workloads and evolving market needs
• Practical examples of how an AI SDK streamlines model deployment and optimization across AI applications