Exercise 01: Build a serverless, AI RAG application using data from Azure Cosmos DB
Scenario
Cosmic Works Bike Company (CWBC) is a leading provider of bikes and e-bikes, headquartered in the Netherlands. With a workforce of 1,000-9,999 employees, CWBC has been at the forefront of creating innovative and reliable bikes for a global customer base. Riders today, however, expect more than just a well-crafted bike-they want intelligent guidance, be it for choosing the right bike, maintaining it properly, or finding the best routes.
By integrating AI and cloud services, CWBC created an assistant that answers typical user questions (like “How do I maintain my disc brakes?”) alongside dynamic route recommendations. A retrieval augmented generation (RAG) approach helps the assistant pull accurate data from Cosmos DB, ensuring it only references real CWBC products and information.
In this exercise, you’ll build serverless, AI RAG applications using .NET Aspire, Semantic Kernel, and Azure Cosmos DB with DiskANN and vector search. You’ll implement Semantic Kernel Azure OpenAI extensions, NoSQL connectors, and semantic caching. You’ll gain practical insights into how to design, build, deploy, and scale RAG pattern applications in Azure. The skills in this lab will give you a solid foundation to create your own generative AI applications. It’s recommended to have C# or other programming language experience before completing this lab.
Objectives
After completing this lab, you’ll be able to:
- Build serverless, AI RAG applications using .NET Aspire, Semantic Kernel, and Azure Cosmos DB with DiskANN and vector search.
- Design, build, deploy, and scale RAG pattern applications in Azure.
Duration
Estimated time: 150 minutes