Do you have customers asking questions like “What is the refund policy?”
Do you manage a long-running onboarding process that spans multiple documents and teams?
What happens when employees need quick answers about health benefits, vacation policies, or internal procedures?
In this hands-on 4-hour workshop, you’ll learn how to turn your existing documentation—like Word files stored in Azure Storage—into an intelligent, conversational experience. Using Azure OpenAI and Azure AI Search, you’ll build a secure, private chat system that retrieves accurate answers from your data in real time, without requiring human intervention.
What You Will Learn:
• How to use Retrieval-Augmented Generation (RAG) to enable natural language queries over your internal data
• How to ingest and index Word documents from Azure Storage using Azure AI Search
• How to create and store vector embeddings for semantic search and relevance ranking
• How to structure and chunk content to improve accuracy and user experience
• How to secure access to private data while enabling real-time chat interaction
• How to refine prompts to optimize responses from Azure OpenAI
Your Takeaways:
• A working prototype that allows users to ask natural-language questions and get accurate, document-sourced answers
• A repeatable pattern for building secure RAG-based systems using Azure tools
• Hands-on experience integrating Azure OpenAI, AI Search, and Storage into a cohesive solution
• A launchpad for internal knowledge portals, support bots, or customer-facing AI assistants
• Confidence to apply generative AI to real-world scenarios in your organization
This workshop is ideal for developers, architects, and technical leaders looking to bring conversational AI into their workflows using secure, enterprise-grade Azure services.

We will provide access to the azure resources needed to complete the hands-on lab as long as you have a PERSONAL (not work) email address and you get it to us before the start of the workshop. Or you can also use your own personal subscription.