[EFFICIENT MODEL DEPLOYMENT & SERVING TRACK]: A Software-First Approach to AI at the Edge | Kisaco Research

Flexible and programmable solutions are the key to delivering high performance, high efficiency AI at the edge. As semiconductor technologies experience the biggest shift in decades in order to meet the requirements of the latest generation of AI models, software is set to be the true enabler of success. Optimised libraries and toolkits empower all stakeholders in the developer journey to follow the “functional to performant to optimal” workflow typical of today’s edge compute application development cycle.

This presentation will present a software-first approach to enabling AI at the edge, touching on the importance of community-wide initiatives such as The UXL Foundation, of which Imagination is a founding member. 

Imagination is a global leader in innovative edge technology, delivering landmark GPU, CPU and AI semiconductor solutions across automotive, mobile, consumer and desktop markets for over thirty years. 

Session Topics: 
Edge
Infrastructure
MLOps
Speaker(s): 

Author:

Tim Mamtora

Chief of Innovation and Engineering
Imagination

Tim Mamtora

Chief of Innovation and Engineering
Imagination