As AI services and applications grow exponentially, including the latest burgeoning GenAI trend, the industry is rapidly transitioning to a Hybrid AI model that splits the compute needed between the Cloud and an Edge device. The need for specialized AI processing on edge devices is becoming almost mandatory. This is driven by a multitude of factors including speech and visual processing for improved Human Machine Interfaces, time-series, or signal processing for applications like vital signs prediction in healthcare and preventative maintenance in factory automation, just to name a few, along with the general need for privacy and security. To meet portability, responsiveness, and cost requirements, future proofed AIoT devices will need to be equipped with efficient compute for advanced sequence prediction, semantic segmentation, multimedia processing, and multi-dimensional time series processing in a very low power budget and silicon footprint.
Nandan Nayampally
Nandan is an entrepreneurial executive with over 25 years of success in building or growing disruptive businesses with industry-wide impact. Nandan was most recently at Amazon leading the delivery of Alexa AI tools for Echo, Fire TV and other consumer devices. Prior to that he spent more than 15 years at Arm Inc. including roles as GM developing Arm’s CPU and broader IP portfolio into an industry leader that is built into over 100B chips. He started his career at AMD on their very successful Athlon processor program. He also helped grow product lines in startup businesses such as Silicon Metrics and Denali Software which had successful acquisitions down the road.
BrainChip
Website: https://brainchip.com/
BrainChip is the worldwide leader in Edge AI on-chip processing and learning. The company’s first-to-market, fully digital, event-based AI processor, Akida™, uses neuromorphic principles to mimic the human brain, analyzing only essential sensor inputs at the point of acquisition and processing data with unmatched efficiency, precision, and energy economy.BrainChip’s Temporal Event-based Neural Networks (TENNs) build on State-Space Models (SSMs) with time-sensitive, event-driven frameworks that are ideal for real-time streaming applications. These innovations make low-power Edge AI deployable across industries such as aerospace, autonomous vehicles, robotics, industrial IoT, consumer devices, and wearables. BrainChip is advancing the future of intelligent computing, bringing AI closer to the sensor and closer to real-time.
