AIoT Ops - Using Kubernetes and ML Ops to Build Edge ML Applications

Explore the innovative world of AIoT (Artificial Intelligence of Things) in this 45-minute conference talk from KubeCon + CloudNativeCon Europe 2022. Discover how to leverage Kubernetes and ML Ops to build cutting-edge Edge ML applications. Delve into the advantages of bringing machine intelligence closer to the edge, particularly for IIoT and IoMT solutions requiring closed-loop decisions. Learn about a novel approach to AIoT application development using Kubernetes, and gain insights into emergent AIoT behaviors and architecture patterns. Witness a practical demonstration simulating an industrial IoT environment, showcasing real-time sensor data processing on an edge device with TPU acceleration for predictive maintenance. Understand how ML pipelines measure drift, re-train, and re-deploy models in AIoT contexts. Gain valuable knowledge about choosing appropriate open source projects for AIoT solutions and implementing effective workflow pipelines.

Introduction
What is AIoT
Reference Architecture
Reference Implementation
Demo
Workflow Pipelines
Demo Setup
Demo Walkthrough
Model Registry
Demonstration
Questions