Exercise - Deploy copilot template for store operations and the necessary Azure resources

  • 10 minutes

In this exercise, you assume the role of Oscar and perform the following tasks:

Task - Deploy the AI store operations package

In this task, you assume the role of Oscar Ward and deploy the AI store operations package in your environment with the following steps.

  1. Go to Microsoft Cloud Solution Center to start the deployment process. Select the Retail cloud from the left pane.

    Screenshot of Retail in the Industry Clouds area of the left pane.

  2. In the Copilot template for store operations tile, select Quick view to learn more about the capability, solutions, prerequisites, and services that are part of the feature. When you’re done, select Close.

    Screenshot of the Copilot template for store operations quick view.

    Screenshot of the Close button.

  3. Select the Add checkbox in the Copilot template for store operations tile and then select Deploy. A workflow to deploy the solution should display.

    Screenshot of the Add checkbox selected.

  4. On the Step 1 Additional components page, select the checkbox near AI Store Assistant ARM Template and then select Next.

    Screenshot of the additional components step.

    Important

    If you don’t select the ARM template at this point, you need to manually deploy the Azure Cognitive Search, Azure Blob Storage Container, and Azure AI Search resources later.

  5. Select MCRIAD, the Microsoft Power Platform environment where you deployed the Store Operations Assist solution, and then select Next.

    Screenshot of the Select a Power Platform environment step.

  6. In Step 3: Select an Azure environment for copilot template for store operations, provide the following details and then select Next.

    • Subscription - Select the Azure Subscription with Azure OpenAI Service turned on.

    • Resource Group - Select Create new. For the Name, enter mcriadRG, and for Region, enter East US. Select Create.

      Screenshot of the Resource Group details.

    • Resource Base Name - mcr

    • Location - East US

    Screenshot of the Select an Azure environment step.

  7. Check the deployment summary and validate it. Read and provide consent to the Terms of service and AI preview terms by selecting the respective checkboxes. Select Next.

    Screenshot of the deployment summary step.

  8. In this step, the workflow checks for dependencies. If Store Operations Assist is already installed, a Configured message with a green check mark displays. Select Deploy.

    Screenshot of the required configurations.

The deployment process initiates automatically, and the program displays its progress. Deployment might take some time (30 minutes in some instances). Afterward, the successful or failed deployment message displays.

Screenshot of the deployment success message.

Congratulations, you successfully deployed copilot template for store operations, or AI store operations, for Store Operations Assist.

Task - Validate resources that you deploy through the Solution Center

In this next task, you validate resources that you deploy through Microsoft Cloud Solution Center.

  1. To validate the deployment in Azure portal, sign in with your admin credentials.

  2. In the upper bar, search for and select Subscription. On this page, Subscription ID is available, which you use in the later exercises.

  3. Select Resource groups.

    Screenshot of the Resource groups button.

  4. Go to the mcriadRG resource group. On the Resources tab, three different resources for Azure Cognitive Search, Azure OpenAI, and Blob Storage account are available.

    Screenshot of the resource group.

    Important

    Note the values of the resources. You need these values to link Store Operations Assist with Azure OpenAI accounts.

  5. Select the mcropen-ai Azure OpenAI resource and then select Go to Azure OpenAI Studio.

    Screenshot of the Go to Azure Open AI Studio button.

  6. Validate that the program automatically deploys all three models in Azure AI Studio. Select Deployments to view the three models: gpt35 turbo, gpt35-turbo-16k, and embeddings.

    Screenshot of the deployments.

    Important

    Note the values of the models. You need these values to link Store Operations Assist with Azure OpenAI accounts.

Reference names to use in further exercises

You use the Resource base name that you previously provided as a prefix for the deployed resources/models. The program adds the prefix by default to the following resources.

Azure Resources

  • Search service - <prefix>cognitivesearch

  • Azure OpenAI - <prefix>open-ai

  • Storage account - <prefix>qnastorage

  • Container - <prefix>qnafiles

Azure OpenAI deployments

  • gpt-35-turbo - <prefix>gpt-35-turbo

  • gpt-35-turbo-16k - <prefix>gpt-35-turbo-16k

  • text-embedding-ada-002 - <prefix>text-embeddings