From Model to Model as a Service: Building Scalable AI Inference

Welcome

In this hands-on lab you take a model through the full serving lifecycle on Red Hat AI Enterprise — from deployment and evaluation through Model-as-a-Service exposure, application integration, and observability.

Duration

~2 hours (including ~10 min intro)

Audience

AI engineers, platform engineers, and AI operations teams with basic OpenShift/Kubernetes knowledge and familiarity with AI model concepts.

What you will do

Module Title Duration

1

Deploy a Model on Red Hat AI Enterprise

10 min

2

Evaluate and Benchmark Model Performance with EvalHub

20 min

3

Expose the Model as a Model-as-a-Service

20 min

4

Integrate the Model into an Application

20 min

5

Validate the End-to-End Workflow

20 min

6

Monitor Model Performance through Observability

20 min

Prerequisites

  • Basic understanding of containers and Kubernetes/OpenShift

  • Familiarity with generative AI model concepts (what a model is, what inference means)

Environment

Your lab environment is a Red Hat AI Enterprise cluster. No models are pre-deployed — you build up the full serving lifecycle from scratch.

Start: Getting Connected → to log in and verify your environment before starting Module 1.