AI/ML Lifecycle Automation at the Edge

Welcome to the AI/ML Lifecycle Automation at the Edge workshop!

Explore how AI and edge computing work together to solve real-world industrial automation challenges with Red Hat’s edge platforms and AI/ML technologies.

About this workshop

This hands-on workshop demonstrates an end-to-end AI solution for battery health monitoring in autonomous transportation robots operating in industrial facilities. You’ll explore the following technologies working together to enable intelligent automation at the edge:

  • Red Hat Device Edge (MicroShift) - Running on an autonomous transportation robot

  • Single Node OpenShift (SNO) - Plant edge server for model training

  • Red Hat OpenShift AI - Complete AI/ML platform on SNO

The lab covers GitOps deployment, AI inference, model serving and automated model training via pipelines. You’ll see how AI models can run at the edge while continuously improving without manual intervention.

Part of the infrastructure is pre-configured. You’ll review configurations and execute hands-on tasks to understand the complete AI/ML lifecycle.

Who this is for

This workshop is designed for platform engineers, ML engineers, DevOps teams, and data scientists who want to:

  • Deploy AI/ML workloads on edge platforms with limited resources

  • Implement automated MLOps workflows for continuous model improvement

  • Work with Kubernetes-based edge infrastructure (MicroShift, Single Node OpenShift)

  • Use Red Hat OpenShift AI for the complete ML lifecycle

  • Build solutions for IoT, industrial automation, or edge AI use cases

What you’ll learn

By the end of this workshop, you will understand how to:

  • Deploy AI/ML workloads on edge platforms (MicroShift, SNO)

  • Use Red Hat OpenShift AI components (workbenches, model serving, pipelines)

  • Implement automated MLOps workflows for continuous model improvement

  • Configure model serving architectures for edge inference

  • Enable pipeline automation for hands-off model retraining

  • Configure alerting and monitoring for critical edge components

Prerequisites

Before starting this workshop, you should have:

  • Basic Kubernetes/OpenShift concepts (pods, deployments, services)

  • Basic Python and Jupyter notebooks knowledge

  • Basic machine learning concepts (training, inference, model validation)

  • Experience with command-line tools (ssh, oc)

Lab modules

This workshop consists of 6 hands-on modules:

Introduction - Business scenario and solution architecture

Technical Details - Environment specs and lab access

Module 1: Transportation Robot

  • Robot infrastructure setup

  • MinIO storage deployment

  • Model serving deployment

  • Battery monitoring system deployment

Module 2: Red Hat OpenShift AI configuration

  • Review Red Hat OpenShift AI installation

  • Review data science project

  • Review data connection to robot storage

Module 3: Model training

  • Review training workbench

  • Import training notebooks

  • Train battery health models

Module 4: Model serving

  • Deploy stress detection model server

  • Deploy time-to-failure model server

  • Query inference endpoints

Module 5: Pipeline automation

  • Review pipeline server

  • Execute automated retraining pipeline

  • Schedule continuous automation

Module 6: Test alerts

  • Review BMS dashboard

  • Test battery health alerts

Conclusion and next steps - Summary and next steps

Estimated time

Total duration: Approximately 1 hour

Let’s get started!

Ready to explore AI/ML automation at the edge?

Navigate to Introduction to begin the workshop.