Azure OpenAI RAG workshop - Node.js version
A hands-on workshop sample that builds a retrieval-augmented-generation chat application with LangChain.js and OpenAI, deployable to Azure Static Web Apps, Azure Container Apps and Azure AI Search.
This sample builds a Node.js chatbot with Fastify, the OpenAI SDK and LangChain.js that answers questions grounded in a corpus of documents, backed by Azure AI Search as the vector database, with a website for interacting with the API. It is structured as a monorepo with all package source under one src folder, and is meant as a starting point for more complex AI applications as well as a companion to a full-length workshop that walks through how it was built.
Deployment uses the Azure Developer CLI: logging into Azure and running one command provisions the needed Azure resources, packages the services, and deploys them; a separate command tears everything down afterward. It can also be opened directly in GitHub Codespaces, or run locally with Docker and the VS Code Dev Containers extension.
Deploying it requires an Azure account with access enabled for the Azure OpenAI service and role-assignment permissions on the target subscription or resource group. It is MIT licensed.