First, setup the CosmosDB NoSQL and Azure Function solution (azure-function-nodejs) to create and store embedding vectors in CosmosDB. Next, setup and run the API application (nodejs) to perform RAG-based semantic search on the Azure CosmosDB NoSQL with embedding vectors-based search for user prompts.
This Azure function will be triggered to create and store embedding vectors in CosmosDB once a file is uploaded to Azure Blob Storage or local storage emulation.
This sample API application demonstrates how to perform RAG-based semantic search on the Azure CosmosDB NoSQL with embedding vectors-based search for user prompts.