Guided project: Build and Deploy your own AI Copilot

  • Module
  • 7 Units
Intermediate
Data Engineer
Developer
Azure Cosmos DB
Azure Kubernetes Service (AKS)
Azure OpenAI Service

Build an AI copilot for Contoso Bike Shop. Integrate Azure OpenAI and Cosmos DB to enable advanced data searches, use GPT-3.5 for enhanced results, and deploy the solution on Azure Kubernetes Service.

Learning objectives

  • Load data from Azure Blob Storage into a local directory and import it into vCore-based Azure Cosmos DB for MongoDB, generating embeddings for product fields.

  • Create a vector index on the generated embeddings and perform vector searches.

  • Enhance search results using GPT-3.5

  • Deploy the application on Azure Kubernetes Service.

Prerequisites

  • Access to an Azure subscription.

  • Access to Azure OpenAI studio.

  • Familiarity with Azure and the Azure portal.

  • Experience programming with Node.js