Introduction

The repository contains a prototype application that explores Azure AI Foundry and its SDK.

It has an Azure DevOps CI/CD pipeline that:

• Builds a containerized Blazor application published in an Azure Container Registry;
• Uses Terraform to plan the resources to create in an Azure subscription;
• Uses Terraform to deploy the containerized application in an Azure Container Application;

Regarding Azure AI Foundry, the repository creates also the following resources:

• An Azure AI Service;
• An Azure AI Foundry;
• An Azure AI Foundry Project;
• A chat model GPT-4.1-nano (careful regarding its location as there are severe quotas)
• Two text-2-speech models:
  • GPT-4o-mini-tts (careful regarding its location - EastUS2 for now in North America, as there are severe restrictions)
  • Eleven Labs

Setup

The requirements are:

• An Azure Container Registry resource where to store the container images. It is not part of the terraformed resources;
• A Resource Group and a Storage Account where to store the Terraform state file. It is not part of the terraformed resources;

Finally, the Azure DevOps pipelines requires the following service connections set up:

• containerRegistry_ServiceConnectionName: a Docker connection to the ACR repository;
• acrResourceGroup_ServiceConnectionName: an ARM connection to the ACR repository, used to clean-up outdated images;
• prototypesSubscription_ServiceConnectionName: an ARM connection to the Azure subscription where the service is deployed