Azure Cosmos DB is a distributed NoSQL database providing unmatched availability, performance and scalability. Azure Cosmos DB is used by OpenAI to store all the chat history for ChatGPT allowing it to grow to 100 million daily active users in record time. This performance is now combined with the new vector capabilities of DiskANN developed by Microsoft Research. This makes it the best database service in Azure for building Intelligent, AI-enabled applications with industry-leading performance and scale.

In this session, we will explore how to build rea-time AI-enabled applications using Azure Cosmos DB. This includes how to implement RAG pattern in C# with vectorized data along-side transactional data in Azure Cosmos DB demonstrating both vector and hybrid search techniques over data with extreme throughput and concurrency. We will also demonstrate how to design and implement chat history using Cosmos DB and demonstrate how you can vectorize this data as well to analyze and improve the performance, cost and accuracy of your RAG Pattern apps.

This will be first announced at BUILD 2024. Target audience are developers.