Azure OpenAI Service for Cloud Native Applications

Chapter 3. Implementing Cloud Native Generative AI with Azure OpenAI Service

This chapter will focus on the implementation of generative AI architectures with Microsoft Azure and Azure OpenAI models, always aiming to present all available options, and minimize the required development, integration, and usage cost, while accelerating the operationalization. For that purpose, I’ve included a series of best practices and typical architectures that will allow you to choose the best building blocks for your specific scenarios.

We will include the most relevant Azure OpenAI implementation approaches, based on existing features and repositories that will continue evolving, improving, and including new functionalities. I’ve included URLs to the original documentation because they are continuously updated with new features, so these links will allow you to explore any details you need. Most of them rely on official accelerators from GitHub repositories, and projects that you will be able to follow and/or fork. But before getting into the details, let’s explore some fundamental topics that will help you understand the full extent of what a generative AI with Azure OpenAI Service means.

Defining the Knowledge Scope of Azure OpenAI Service–Enabled Apps

Generative AI applications on Microsoft Azure are not only for regular ChatGPT-type applications. They are advanced architectures that rely on diverse technology pieces, including the core infrastructure (servers, GPUs, etc.) required to ...