chat-with-your-data-solution-accelerator
A Solution Accelerator for the RAG pattern running in Azure, using Azure AI Search for retrieval and Azure OpenAI large language models to power ChatGPT-style a
What is chat-with-your-data-solution-accelerator?
A Solution Accelerator for the RAG pattern running in Azure, using Azure AI Search for retrieval and Azure OpenAI large language models to power ChatGPT-style and Q&A experiences. This includes most common requirements and best practices.
Overview of chat-with-your-data-solution-accelerator
What it is
Chat with Your Data is a solution accelerator for the RAG (Retrieval-Augmented Generation) pattern running in Azure. It uses Azure AI Search for retrieval and Azure OpenAI large language models to power ChatGPT-style and Q&A experiences. The solution indexes unstructured content from contracts, policies, product manuals, and benefit guides, and puts a natural-language chat experience in front of it. It deploys into an Azure subscription with a single azd up command.
Key features
- Responses are grounded in indexed content with inline citations back to the source documents.
- Upload files or index public web pages through a document ingestion pipeline that parses, chunks, and embeds many supported file types.
- Choose Azure AI Search with Cosmos DB or PostgreSQL with pgvector as the retrieval and persistence engine, selected at deploy time.
- Choose between two interchangeable orchestrators, Agent Framework or LangGraph, that share the same retrieval and grounding pipeline and are selected at deploy time.
- A collapsible reasoning panel shows the model's intermediate steps alongside the answer as they stream in.
- Azure AI Content Safety screens prompts and responses to help keep the experience within policy boundaries.
- A single user-assigned managed identity and Azure RBAC authorize every downstream call, with no Key Vault and no application secrets to manage.
Use cases
Organizations hold vast unstructured knowledge in contracts, policies, product manuals, and benefit guides that is hard to search and slow to answer questions from. This solution indexes that content and provides grounded answers with inline citations, so people find answers in seconds instead of digging through files.
Getting started
Follow the quick deploy steps on the deployment guide to deploy this solution to your own Azure subscription. Click here to launch the deployment guide.
License and pricing
The license is MIT. Pricing varies by region and usage; use the Azure pricing calculator to estimate costs. Resources continue to incur charges until deleted.
Based on the project's public README and repository metadata.
