Skip to content

contentexp.md

Latest commit

 

History

History
31 lines (16 loc) · 4.18 KB

contentexp.md

File metadata and controls

31 lines (16 loc) · 4.18 KB

Content and Experiences

Image

  • AI Kick Off Projects: This is a collection of AI-focused Challenge Project modules. Each module will enable you to build an AI-empowered end-to-end solution based on a specification. This collection will be regularly updated with new projects, so keep an eye out for the latest content.

  • Understanding the difference in using different large langauge models: In this blog, you will learn how to leverage different Large Language Models available on Azure Machine Learning and provided by Hugging Face, Azure ML, OpenAI, and Meta. Also, you will learn how to integrate them into your Web Application or Power App.

  • Let's build a PlugIn for ChatGPT and BingChat: In this workshop, we will create a plugin for ChatGPT from scratch. We will see examples in different programming languages such as Javascript, .NET, or Python. We will also explore the architecture that supports a plugin and how this model can be used for other platforms like BingChat.

  • Tips and Tricks to Bring Your AI Idea to Life: Join us as we explore the powerful tools and technologies from Microsoft that can drive the development of your AI projects. In this weekly series, we'll be inviting experts to share their experiences and insights on how to turn your ideas into practical and innovative solutions.

  • Prompt flow in Large Language Models How do you better integrate code and prompts into an application? Prompt flow in Azure AI Studio is the answer. This session will introduce the concept of prompt flow and related usage skills, so that developers can better build enterprise-level Copilot applications

  • A 'Lets get the best of Open AI in ... in Power Platform, in Power Automate and in Power Virtual Agents: If you want to know how to add all of this power to Power Platform apps, these sessions are for you. They add OpenAI GPT capabilities to a couple of Power Platform End-to-End solutions.

OpenAI models and Azure integration

Do you wish to integrate Large Foundation Models - like OpenAI models - with Azure services?

Here's some tips and tricks

  • Use Azure SDKs when available: for example the Javascript SDK, the Python SDK and the .NET SDK enable you to use OpenAI models from your code and to switch from an Azure OpenAI endpoint to a non-Azure OpenAI endpoint with minimal code changes.

  • Explore the OpenAI models available in the Foundation Models catalogue in Azure Machine Learning Studio. You can handle fine-tuning, deployment and the whole machine learning lifecycle from there.

  • Use an AI orchestrator, like Semantic Kernel or LangChain to combine multiple AI services with conventional programming languages (like C# and Python) and add your own logic.

  • Leverage Azure PromptFlow to make it easier to debug, test and compare different models performances and collaborate on different project components.

  • Deploy an Azure Content AI Safety component to filter inputs and outputs of Large Foundation Models, mitigating the risk of manipulation and harmful content generation.