Developing AI-Powered Apps with Azure AI Services

What is Artificial Intelligence?

In this chapter you will get a short overview about what AI is exactly, and what we can do with it.

• Definitions of Artificial Intelligence
• Domains of Artificial Intelligence
• History, Current State and Future

Introduction to Azure AI Studio and Azure AI Services

Your intelligent app needs to understand its environment and make decisions. For that you can use pre-trained Azure AI Services that detect sentiment, recognize speakers, understand pictures, etc. Azure AI Studio is a web portal that brings together these services into a single, unified development environment. Azure AI services provides a comprehensive suite of out-of-the-box and customizable AI tools, APIs, and models that help modernize your business processes faster.

• What is Azure AI Studio?
• Creating Azure AI Services in Azure AI Studio
• Accessing Azure AI Services
• Collaborating efficiently and effectively on AI projects

Azure AI Language

Your application would be useless if they didn't understand their users. Azure AI Language allows us to create models to interpret what the user wants, as well as extract key information from natural language.

• Conversational Language Understanding
• Intents, Entities and Utterances
• Key Phrase Extraction
• Named Entity Recognition
• Language Detection
• Sentiment Analysis and Opinion Mining
• LAB: Adding Natural Language Processing to the CSI Bot

Azure AI Vision

Azure's Azure AI Vision service gives you access to advanced algorithms that process images and return information based on the visual features you're interested in.

• Getting Started with Azure AI Vision
• Optical Character Recognition (OCR)
• Face Detection and Face Recognition
• Azure AI Custom Vision
• LAB: License Plate Detection

Azure AI Document Intelligence

AI Document Intelligence is an AI service that applies advanced machine learning to extract text, key-value pairs, tables, and structures from documents automatically and accurately. Turn documents into usable data and shift your focus to acting on information rather than compiling it.

• What is Azure AI Document Intelligence ?
• Using Document Intelligence Studio
• Working with Document Analysis Models
• Creating a Custom Model
• Using Models in Code
• LAB: Processing documents using Azure AI Document Intelligence

Azure OpenAI and ChatGPT Fundamentals

This module introduces Azure OpenAI and ChatGPT Large Language Models (LLMs). Learn about the available LLM models, how to configure and use them in the Azure Portal. Additionally you will learn about the Transformer architecture, the succesful neural network architecture behind LLMs ChatGPT-like GPT-4.

• Introducing Azure OpenAI and ChatGPT
• The Transformer Model and Embeddings
• Azure OpenAI Studio
• Configuring Deployments
• ChatGPT, GPT Models and Large Language Models
• Understanding Tokens
• LLM Pricing
• Role Management: System, User and Assistant
• Guide to Effective Prompt Engineering
• Zero, One- and Few-Shot Prompting
• LAB: Deploying and Using Azure OpenAI

The Chat Completions API

In this module, we'll delve into the APIs provided by ChatGPT and Azure OpenAI. You'll discover how to seamlessly integrate them into your applications. The latest GPT models introduce Function Calling, a new capability to connect large language models to external tools, services or code. This enables the creation of your own smart, AI powered Copilots.

• An introduction to the Chat Completions API
• Azure OpenAI SDK
• OpenAI Function Calling Explained
• LAB: Discovering the Chat Completions API

Using your own data in a LLM with Azure AI Search

Azure AI Search facilitates the adoption of the Retrieval Augmented Generation (RAG) design pattern. This methodology involves retrieving pertinent information from a data source and leveraging it to enhance the output of generative AI models.This symbiosis between retrieval and generation sets a new standard for AI-driven search solutions.

• What is Azure AI Search?
• Retrieval Augmented Generation
• Creating an Index on your Own Data
• Introduction to Embeddings and Vector Search
• AI Enrichment with your own Data
• Using the Azure OpenAI SDK
• Privacy Concerns
• Fine-tuning vs RAG
• LAB: Chat with Azure OpenAI models using your own data

Orchestrating AI Models using Semantic Kernel

The Semantic Kernel is an open-source SDK backed by Microsoft that seamlessly integrates Large Language Models such as OpenAI and Azure OpenAI with programming languages like C#. It allows users to use natural language input within Large Language Models to seamlessly invoke and interact with your custom code.

• An Introduction to Semantic Kernel
• Integrating LLMs in your applications
• Keeping track of Token Usage
• Enable AI Models to execute code using Plugins
• Solve complex problems through use of Planners
• LAB: Create a Natural Language to SQL Translation Copilot