Creating Generative AI Solutions and Copilots with Azure AI Foundry (Formerly Azure AI Studio)
1h 46mIntermediate2024-12-13
Authors
Rodrigo Díaz Concha
Solutions Architect, Microsoft Regional Director, and Microsoft MVP
Course details
Explore the world of generative AI applications and master the tools and techniques necessary to create robust and innovative solutions using Azure AI Studio—a unified platform for developing, testing, and deploying intelligent AI solutions. Learn how to build production-ready copilot applications that take advantage of cutting-edge large language models (LLMs) and manage the entire lifecycle, from concept and design to deployment and monitoring.
Learning objectives
Explain the differences between Azure AI Studio and other offerings such as Azure Open AI Studio.
Be able to recommend a large language model (LLM) for a given scenario or context.
Create a copilot using Azure AI Studio.
Create a fine-tuned model from a foundational model and additional data.
Learning objectives
Explain the differences between Azure AI Studio and other offerings such as Azure Open AI Studio.
Be able to recommend a large language model (LLM) for a given scenario or context.
Create a copilot using Azure AI Studio.
Create a fine-tuned model from a foundational model and additional data.
Skills covered
Azure AI ServicesGenerative AIAzureSoftware Development ToolsArtificial Intelligence (AI)MicrosoftSoftware DevelopmentDeep Dive (X:Y)
Concepts
0. Introduction
- 01 - Welcome to Azure AI Foundry Azure AI Studio
1. Introduction to Azure AI Foundry Azure AI Studio
- 02 - What is Azure AI Foundry Azure AI Studio
- 03 - What you need to know
- 04 - Setting up your Azure AI Foundry Azure AI Studio account
- 05 - Navigating the Azure AI Foundry Azure AI Studio portal
- 06 - Exploring the model catalog
- 07 - Understanding the model benchmarks
- 08 - Exploring additional AI services
2. Getting Started with Hubs in Azure AI Foundry Azure AI Studio
- 09 - Creating a hub using the Azure AI Foundry Azure AI Studio portal
- 10 - Creating a hub using the Azure portal
- 11 - Creating a hub using the Azure CLI
- 12 - Managing connected resources for the hub
- 13 - Deploying an OpenAI model to a hub
- 14 - Managing content filters
- 15 - Managing users for the hub
3. Creating Projects and Using the Playgrounds in Azure AI Foundry Azure AI Studio
- 16 - Creating your first project
- 17 - Deploying a model using the models as a service feature
- 18 - Using the assistants playground
- 19 - Using the images playground
- 20 - Using the completions playground
- 21 - Grounding the model response with the add your data feature
- 22 - Incorporating a Copilot into a Python application
4. Building Your First Copilot with Prompt Flow in Azure AI Foundry Azure AI Studio
- 23 - Creating your first flow
- 24 - Understanding nodes and tools in a flow
- 25 - Configuring the tools and testing the flow
- 26 - Deploying the flow to an endpoint
- 27 - Testing the endpoint using Postman
- 28 - Updating the flow and creating a new deployment
5. Building a Chat Copilot with Prompt Flow and Azure Cosmos DB in Azure AI Foundry Azure AI Studio
- 29 - Provisioning the Azure Cosmos DB resource and loading the data
- 30 - Creating the chat flow
- 31 - Implementing the code for connecting to Azure Cosmos DB
- 32 - Augmenting the prompt using the Jinja format
- 33 - Testing the endpoint using the portal and Postman
Conclusion
- 34 - Resources for additional learning