Expedera Inc.

In an "Ask the Expert" webinar with Rambus, Sharad Chole, co-founder and chief scientist at Expedera Inc., discusses the rapid evolution of artificial intelligence (AI) and how it is transforming edge computing. Sharad shared his insights on the challenges and opportunities of deploying AI inference workloads at the edge. https://lnkd.in/gQFbMSDH

Advancements like this will be critical to ensuring the future edge AI, specifically LLMs. With model sizes and memory requirements being the biggest barrier for deployment in battery-powered devices, 56% reductions in model size and a 41% reduction in memory usage point towards a future where LLMs can cost- and power-effectively be deployed in your smartphones, cars, notebooks, and other consumer devices. https://lnkd.in/gbqf2TCN

AI is inspiring a lot of creativity in chip architecture today, which Ann Mutschler from Semiconductor Engineering writes about today. The article addresses the fact that the number of approaches to AI inferencing is growing to deal with unique applications and larger & more complex models. Expedera comments on this and more! https://ow.ly/McXi50TNEu4

ICYMI: Semiconductor Engineering explores how AI is bringing the problem of connecting different data types and processes to the forefront

The Origin E6 is a versatile NPU IP family customized to match the needs of next-generation smartphones, automobiles, and consumer devices. With support for video, audio, and text-based AI networks, including standard, custom, and proprietary networks, the E6 is the ideal hardware/software co-designed platform for chip architects and AI developers. It offers broad native support for current and emerging AI models, and achieves ultra-efficient workload scheduling and memory management, with up to 90% processor utilization—avoiding dark silicon waste. Learn more: https://lnkd.in/gc2FMAVT

Expedera Inc.'s Paul Karazuba will be moderating Thursday's Global Semiconductor Alliance webinar, "From a Venture Capitalist's Perspective", focused on VC investment in the semiconductor industry with a panel including Steve Fu, Andrew Byrnes, and Mark Yusko. Registration is free via the link in the post below. Join us!

On-device AI is a requirement for most, if not all, new chip designs. Silicon architects need solutions that support the latest AI models with the best balance of power and area requirements. Expedera's E2 family is ideal for chips destined for smartphones, consumer devices, home appliances, and other applications. Learn more here: https://lnkd.in/eQzSif_Z

While AI holds the potential to change how companies interact throughout the global semiconductor ecosystem, gluing together different data types and processes that can be shared between companies that in the past had little or no direct connections. This is explored in a special report from Semiconductor Engineering, with comments from Expedera Inc.'s Patrick Donnelly. https://lnkd.in/gCMtsnSy

In a recent article, Global Corporate Venturing's examines how AI, corporate venture capital, and the semiconductor ecosystem have come together, including comments from Expedera Inc.'s Paul Karazuba https://lnkd.in/e_NmZMdj

Protecting AI with robust security measures is not an option but a necessity in the age of AI. Organizations must take proactive approaches in identifying and mitigating security risks to ensure the integrity, reliability, and proper use of AI systems. By implementing the strategies and considerations outlined in this document, semiconductor companies can enhance their AI security posture, protect their assets, and maintain trust in AI-driven solutions to best protect internal, customer, partner, and vendor assets. In conjunction with the Global Semiconductor Alliance, Expedera presents the "Security in AI" white paper. https://lnkd.in/gEyKYfss