RAG chat app with Azure OpenAI and Azure AI Search
A sample application demonstrating a ChatGPT-style, Retrieval-Augmented Generation experience over a company's own documents, using Azure OpenAI and Azure AI Search.
This solution creates a ChatGPT-like frontend experience over a user's own documents using Retrieval-Augmented Generation (RAG). It uses Azure OpenAI Service to access GPT models and Azure AI Search for data indexing and retrieval, with a Python backend and equivalent JavaScript, .NET and Java samples based on the same design.
Features include a multi-turn chat interface that renders citations and the model's thought process, in-UI settings to tweak behavior and experiment with options, indexing and retrieval across many document formats with cloud data ingestion, optional multimodal reasoning over image-heavy documents, optional speech input and output for accessibility, optional automated login through Microsoft Entra, and performance tracing with Application Insights.
The repository includes sample data for a fictitious company so the deployed application can answer questions about benefits, internal policies and job descriptions end to end; the README notes it is built to showcase Azure services and should not be used in production without additional security hardening.