00 Prerequisites

Goal

Set up your local environment and Azure resources so every subsequent lesson runs without blockers.

Estimated time

20 minutes.

Official references

Required tools

Tool	Minimum version	Install
Azure CLI	2.60+	`curl -sL https://aka.ms/InstallAzureCLIDeb \\| sudo bash`
Docker Desktop	24+	docs.docker.com/get-docker
kubectl	1.28+	`az aks install-cli`
Python	3.11+	python.org
Node.js	18+	nodejs.org
Bicep CLI	0.25+	`az bicep install`

Exercise

az login
az account set --subscription "<your-subscription-id>"

Step 2 — Register required providers

az provider register --namespace Microsoft.ContainerService
az provider register --namespace Microsoft.App
az provider register --namespace Microsoft.OperationalInsights

Step 3 — Create a resource group

export RESOURCE_GROUP=rg-container-workshop
export LOCATION=eastus2

az group create --name $RESOURCE_GROUP --location $LOCATION

Step 4 — Create an Azure Container Registry

export ACR_NAME=acrtriage$RANDOM

az acr create \
  --resource-group $RESOURCE_GROUP \
  --name $ACR_NAME \
  --sku Basic

Step 5 — Microsoft Foundry resource and project

Create a Foundry resource with project management enabled, a project, and a model deployment. You can do this via the Microsoft Foundry portal or the CLI:

# Create the Foundry resource with project management
az cognitiveservices account create \
  --name ai-triage \
  --resource-group $RESOURCE_GROUP \
  --kind AIServices \
  --sku S0 \
  --location $LOCATION \
  --allow-project-management

# Assign a stable custom subdomain
az cognitiveservices account update \
  --name ai-triage \
  --resource-group $RESOURCE_GROUP \
  --custom-domain ai-triage

# Create a Foundry project inside the resource
az cognitiveservices account project create \
  --name ai-triage \
  --resource-group $RESOURCE_GROUP \
  --project-name triage-project \
  --location $LOCATION

# Deploy a model
az cognitiveservices account deployment create \
  --name ai-triage \
  --resource-group $RESOURCE_GROUP \
  --deployment-name gpt-4.1-mini \
  --model-name gpt-4.1-mini \
  --model-version "2025-04-14" \
  --model-format OpenAI \
  --sku-name Standard \
  --sku-capacity 30

Tip

If az cognitiveservices account create reports unrecognized arguments: --allow-project-management, run az upgrade and retry.

Copy the project endpoint from the Foundry portal (Home → open your project → the endpoint shown for SDK/API usage):

# Format: https://<your-ai-service>.services.ai.azure.com/api/projects/<your-project-name>

Note

The app uses DefaultAzureCredential for authentication. Make sure you are signed in with az login and have the Cognitive Services OpenAI User role on the Foundry resource.

Step 6 — Configure environment variables

cp .env.example .env

Edit .env and fill in your values:

AZURE_AI_PROJECT_ENDPOINT=https://<your-ai-service>.services.ai.azure.com/api/projects/<your-project-name>
AZURE_AI_MODEL_DEPLOYMENT=gpt-4.1-mini
ACR_NAME=<your-acr-name>
RESOURCE_GROUP=rg-container-workshop
LOCATION=eastus2

Step 7 — Build and push container images

source .env

az acr build --registry $ACR_NAME --image triage-backend:v1 ./app/backend
az acr build --registry $ACR_NAME --image triage-frontend:v1 ./app/frontend
az acr build --registry $ACR_NAME --image triage-worker:v1 ./app/worker

Tip

az acr build builds remotely in Azure, so you don’t need Docker running locally and you avoid Docker Hub rate limits.

Step 8 — Test locally (optional)

# Backend
cd app/backend
pip install -r requirements.txt
source ../../.env
uvicorn main:app --reload --port 8001

# Frontend (new terminal)
cd app/frontend
npm install
npm run dev

Open http://localhost:3000 to see the Patient Triage UI.

Verification

[ ] az account show prints your subscription name.
[ ] docker --version shows 24+.
[ ] kubectl version --client shows 1.28+.
[ ] python --version shows 3.11+.
[ ] node --version shows 18+.
[ ] Resource group rg-container-workshop exists in Azure.
[ ] ACR exists and az acr login succeeds.
[ ] Foundry project endpoint is set in .env.
[ ] All three images are pushed to ACR.