Transform the Retail Experience: Architecting LLM Inferencing Systems for Edge Deployment (Presented by Supermicro)
, Senior Director, Edge AI Solutions, Supermicro, Inc.
As large language models (LLMs) are rapidly integrated into every industry, edge AI is crucial for enabling faster, more secure, and efficient data processing, particularly for applications where real-time analytics and decision-making are essential. We'll explore the transformative potential of deploying powerful systems at the edge to enhance in-store experiences. These systems must be capable of providing high-quality responses with minimal latency. We showcase a retail concierge solution, powered by an LLM inferencing model, that enables customers to have live interactions with an avatar. Next, we provide an overview of designing solutions that address the unique environmental considerations at the edge, including form factor, power delivery, network latency, and serviceability.