Azure OpenAI for Developers
Learn how to integrate OpenAI functionalities and tools in this comprehensive 3-day course.
Course Description
This course offers a comprehensive introduction to AI development with a primary focus on integrating OpenAI functionalities and tools. Designed for developers already familiar with foundational web technologies, this curriculum dives deep into leveraging OpenAI’s capabilities to build AI-powered applications.
Key Highlights:
- Foundation: Begin with an understanding of OpenAI’s key features, capabilities, and potential applications.
- Setting Up: Equip yourself with the required tools and software, including Visual Studio Code, Python, .NET, Git, Postman, NodeJS, and Microsoft C++ Build Tools. Special attention is given to troubleshooting common setup challenges.
- Hands-on Challenges: Transition from theory to practice with seven extensive challenges. These span from provisioning Azure OpenAI resources, understanding API keys and endpoints, integrating various SDKs, building applications using ReactJS and ASP.NET Core, and finally, managing container deployments using Azure services.
- Azure Integration: Learn the intricacies of Azure Container Registry, Azure App Service, and Azure DevOps. Delve deep into the build and release pipelines, ensuring a smooth CI/CD process.
By the end of this course, participants will be adept at creating, testing, and deploying AI-enhanced applications, with a thorough grasp of both OpenAI tools and Azure’s deployment mechanisms.
About this Course
Course Outline
Module 1: Introduction to OpenAI and Large Language Models
- Provisioning Azure OpenAI Resources
- Creating an Azure OpenAI resource
- Create deployment models
- Embeddings
- Chat
- ChatCompletion
- Test the models in the Playground
- Model Settings and Prompt Engineering
- Temperature, Max Tokens, Engine, Stop Sequence
- Prompt Structure:
- System, User and AI Prompts
- Understanding API keys and endpoints
- Using the API in in Postman & cURL
Module 2: Understanding Large Language Models
- What are large language models (e.g., GPT-3) and their applications?
- Hands-on exploration of GPT-3’s capabilities
- Building with Large Language Models
Module 3: Introduction into Azure OpenAI SDK’s
- Azure OpenAI SDKs provide streamlined access to OpenAI’s AI models in Azure.
- Support for popular programming languages, making it accessible to a wide range of developers.
- Examples like JavaScript embedded demo show how to embed AI features in web applications.
- Comprehensive tools and demos available for .NET (C#) developers, similar to Python support.
- Module 4: Using the JavaScript SDK with ReactJS
Module 4: Building a ReactJS application with Azure OpenAI
- Integration of Azure OpenAI JavaScript SDK is essential for interaction with OpenAI models.
- Design a user-friendly interface and use React’s component-based architecture for user interactions with AI.
Module 5: Azure Integration and Advanced Topics
- Integrating OpenAI into Azure
- Overview of Azure’s AI and machine learning services
- How to use Azure Functions and Logic Apps for OpenAI integration
Module 6: Advanced OpenAI Concepts
- Building Advanced OpenAI Applications
Module 7: Deployment, Best Practices, and Use Cases
Module 8: Deployment Strategies
- Choosing the right deployment methods for your OpenAI applications (e.g., Containers, Functions, VMs)
- Scaling and load balancing
- Security, Monitoring, and Best Practices
- Securing OpenAI applications on Azure
- Monitoring, troubleshooting, and best practices
Module 9: Real-World Use Cases and Q&A
- Explore real-world use cases of OpenAI in Azure and have a Q&A session
Challenges
Challenge 1
- Provisioning Azure OpenAI Resources
- Creating an Azure OpenAI resource
- Create deployment models
- Embeddings
- Chat
- ChatCompletion
- Test the models in the Playground
- Model Settings and Prompt Engineering
- Temperature, Max Tokens, Engine, Stop Sequence
- Prompt Structure:
- System, User and AI Prompts
- Understanding API keys and endpoints
- Using the API in in Postman & cURL
Challenge 2
- Introduction to Azure OpenAI SDKs
- Overview of available SDKs
- Javascript embedded demo
- Integrating the C# SDK – demos for AI-050 have a c# and Python option
- Integrating the Python SDK – demos for AI-050 have a c# and Python option
Challenge 3
- Building a Simple ReactJS Application with Azure OpenAI
- Setting up a ReactJS project
- Integrating the Azure OpenAI JavaScript SDK
- Creating user interface
- Testing and debugging
Challenge 4
- Building a Simple ASP.NET Core Application with Azure OpenAI
- Setting up an ASP.NET Core project
- Integrating the Azure OpenAI C# SDK
- Developing AI-powered server-side logic
- Testing and debugging
- Docker image creation
Challenge 5
- Create an Azure Container Registry (ACR) instance.
- Create an Azure App Service instance to host container
- Upload container to ACR
- Deploy container to App Service
Challenge 6
- Create a new project in Azure DevOps
- Set up code repository
- Create a new build pipeline.
- Use the existing Dockerfile to define the build process.
- Add tasks to build and push the Docker image to the Azure Container Registry.
- Trigger the build
Challenge 7
- Create a new release pipeline.
- Add an artifact that points to the build pipeline’s output.
- Define the deployment process.
- Use the deployment center
- Trigger the release
- Validate deployment
Duration
3 Days
Prerequisites
- Visual Studio Code
- Python for Visual Code Studio
- This guide walks you through installing a Python interpreter and the extension needed for using VSCode for Python development.
- Note: Please use version 3.10 of Python. There are issues with existing libraries.
- Upgrade pip to the latest version
- If you are seeing this error “ERROR: Could not build wheels for hnswlib, which is required to install pyproject.toml-based projects” then use the following link as a fix.
- .Net 6 LTS
- Git installed
- Postman
- NodeJS 18+
- Microsoft C++ Build Tools
Level
Advanced
Product
Azure OpenAI
Role
Developer