Azure AI Engineer Associate (AI-102) Study Guide. In-Depth Certification Guide and Practice - Helion

ebook

Autor: Renaldi Gondosubroto
ISBN: 9781098169220
stron: 456, Format: ebook
Data wydania: 2025-09-09
Księgarnia: Helion

Cena książki: 160,65 zł (poprzednio: 198,33 zł)
Oszczędzasz: 19% (-37,68 zł)

Osoby, które kupiły tę książkę, wybierały także »

With the GenAI boom showing no sign of letup, the demand for AI skills will only increase with time and innovation. Microsoft Azure leads the pack with services for developing and deploying AI solutions, so professionals looking to break into this field should consider pursuing certification as an Azure AI Engineer Associate.

Azure's AI-102 exam isn't a piece of cake, but author Renaldi Gondosubroto makes it a great deal more approachable with this comprehensive study guide. Packed with expert guidance, it covers everything you'll need to know to pass the exam. You'll dive deep into all the phases of AI solutions development, from requirements definition and design to development, deployment, and integration, along with maintenance, performance tuning, and monitoring throughout.

The book also takes you through practical implementation of these systems, covering decision support, computer vision, natural language processing, knowledge mining, document intelligence, and generative AI solutions.

Understand the core concepts of Azure AI services
Develop and deploy AI solutions within Azure's environment
Explore integration and security practices with Azure AI services
Optimize and troubleshoot AI models on Azure
Gain knowledge about building GenAI solutions on Azure and put it into practice

Osoby które kupowały "Azure AI Engineer Associate (AI-102) Study Guide. In-Depth Certification Guide and Practice", wybierały także:

Jak zhakowa 125,00 zł, (10,00 zł -92%)
Biologika Sukcesji Pokoleniowej. Sezon 3. Konflikty na terytorium 126,36 zł, (13,90 zł -89%)
Windows Media Center. Domowe centrum rozrywki 66,67 zł, (8,00 zł -88%)
Podręcznik startupu. Budowa wielkiej firmy krok po kroku 92,67 zł, (13,90 zł -85%)
Ruby on Rails. Ćwiczenia 18,75 zł, (3,00 zł -84%)

Spis treści

Azure AI Engineer Associate (AI-102) Study Guide. In-Depth Certification Guide and Practice eBook -- spis treści

Preface
- Why I Wrote This Book
- Who This Book Is For
- How This Book Is Organized
- Conventions Used in This Book
- Using Code Examples
- OReilly Online Learning
- How to Contact Us
- Acknowledgments
1. Introduction to AI Solutions on Microsoft Azure
- Introduction to the AI Landscape
  - The Minimally Qualified Candidate for the AI-102 Exam
  - Where We Are Now with AI
  - Fundamental AI Concepts
    - Data science
    - Machine learning
    - Artificial intelligence
    - Deep learning
  - The Six Core Principles: Considerations for Developing AI Responsibly
  - Explainable AI Techniques
- Tech Setup
  - Setting Up Your Microsoft Azure Account
  - Configuring Your Azure AI Environment
  - Configuring Your Local Development Environment
- Gaining an Understanding of Azure AIs Capabilities
  - The Capabilities of Microsoft Azures AI Services
    - Machine learning
    - Computer vision
  - Consuming AI Services
    - Microsoft Azure user interface
    - SDKs
    - REST APIs
  - Authentication and Security
    - Microsoft Entra ID (formerly known as Azure AD)
    - Subscription keys
    - Managed identities
    - More security measures
  - Billing and Cost Management
- Your Responsibilities as an AI Engineer
  - Meeting Challenges and Managing Risks
    - Technical and scalability challenges
    - Security and compliance
    - Resource availability
    - Ethical and legal issues
  - Continuous Learning and Collaboration
  - User-Centered Design
- Practical: Running a Text Analytics Service
- Chapter Review
- Chapter Quiz
2. Planning and Managing AI Solutions in Microsoft Azure
- The Azure AI Project Lifecycle
  - Requirements definition and design
  - Development
  - Deployment
  - Integration
  - Maintenance
  - Performance Tuning
  - Monitoring
- Practical: Designing an AI Solution
  - A Simple Example of Solution Design
  - Weighing the Trade-offs
- Planning and Configuring Access and Security
  - Implementing the Appropriate Access Control Requirements
    - Data protection
    - RBAC on Azure
    - Managing account keys
    - Working with Azure Key Vault
    - Managing private communications
  - Working with Security over the Network
    - Virtual networks and subnets
    - Network security groups
    - Azure Application Gateway and Web Application Firewall
    - Azure Firewall
    - Networking for individual AI service instances
- Creating and Managing Azure AI Services
  - Deploying an Azure AI Services Resource
  - Cost Optimization Strategies for Azure AI Services
  - Implementing a Container Deployment
  - Working with APIs and SDKs in Azure
    - Standard web service API primary operations
    - Relevant API documentation
    - Construction of the HTTP request
    - Sending requests
    - Handling responses
    - Handling errors
  - Practical: Designing an AI Solution with the REST API
- Monitoring Azure AI Services
  - Proactively Monitoring Costs
    - Microsoft Cost Management
    - Azure Advisor
    - Cost Analysis
  - Using Metrics and Alerts
    - Alerts
    - Configuring diagnostic logging
    - Performance metrics
    - Error metrics
    - Model-specific metrics
  - Practical: Designing Your AI Solution
    - Setting up the key vault
    - Coding the system
    - Monitoring and logging your solution
- Chapter Review
- Chapter Quiz
3. Storing, Interpreting, and Visualizing Data
- Data Storage and Management in Azure AI
  - Choosing Storage Options for AI Solutions
    - Understanding data types and requirements
      - Structured data
      - Unstructured data
    - Performance and scalability
    - Security and compliance
    - Ease of integration
    - Cost
    - Azure storage options
      - Azure Disk Storage
      - Azure Blob Storage
      - Azure Data Lake Storage
      - Azure Files
      - Azure NetApp Files
      - Azure File Sync
      - Azure Stack Edge
      - Azure Data Box
  - Data Management Best Practices
    - Data governance and cataloging
      - Step-by-step guide to cataloging AI data with Microsoft Purview
    - Data quality and preparation
      - Using Azure Data Factory for data preparation
      - Using Azure Databricks for data preparation
    - Data backup and disaster recovery
      - Data backup with Azure Backup
      - Disaster recovery with Azure Site Recovery
      - Implementing and managing backup and recovery strategies
      - Testing and validation
- Data Interpretation for AI Solutions
  - Leveraging Azure AI for Data Analysis
    - Creating event-driven architectures
    - Building a unified solution
  - Model Training and Selection in Azure AI
    - Automated machine learning in Azure Machine Learning
    - Evaluation metrics and tools
      - Evaluation metrics
      - Tools in Azure Machine Learning
    - CI/CD for reproducibility
- Data Visualization Techniques and Real-Time Analytics
  - Introduction to Azure Data Visualization Tools
    - Azure Machine Learning Studio
    - Power BI
    - Azure Data Explorer
    - Additional integrations
  - Real-Time Analytics and Decision Making
    - Real-time analytics
    - Near-real-time analytics
    - Non-real-time analytics
    - Decision framework
- Implementing AI in Data Analysis
  - Integrating AI with Azures Data Platforms
    - Unified data architecture
    - Interoperability and integration
    - Security and compliance
  - Case Study of Data Analysis in Action
- Practical: Building an AI-Powered Analytics Dashboard
  - Setting Up Azure Blob Storage
  - Uploading the Customer Feedback Data
  - Creating an Azure AI Services Language Service
  - Creating and Configuring an Azure SQL Database
  - Setting Up Azure Data Factory
  - Creating a Pipeline for Data Movement and Transformation
  - Creating a SQL Database to Store the Data
  - Visualizing the Data with Power BI
- Chapter Review
- Chapter Quiz
4. Building Decision Support Solutions with Azure AI
- Introduction to Decision Support for Azure AI
  - Understanding Decision Support Systems
  - What Does Decision Support Look Like in the Azure AI Landscape?
- Utilizing Azure AI Metrics Advisor to Implement Data Monitoring Solutions
  - Understanding Azure AI Metrics Advisor
    - Steps to follow when using Metrics Advisor
    - Features
    - When to use it
  - Implementing Data Monitoring Solutions
    - Creating the Azure AI Metrics Advisor resource
    - Connecting data sources
    - Ingesting and preprocessing data
    - Configuring anomaly detection
    - Setting up notifications and alerts
    - Monitoring and diagnostics
    - Continuous improvement
    - Integrating the API
    - Fine-tuning anomaly detection
    - Managing alerts and notifications
    - Configuring notification hooks
    - Querying anomalies and alerts
    - Customizing and extending the monitoring solution
- Text Classification and Moderation with Azure AI Content Safety
  - Understanding Azure AI Content Safety
    - Different types of analysis made possible by Azure AI Content Safety
    - Performance optimization and moderation accuracy
    - Key features
    - When to use it
    - Security
  - Moderating Text with Azure AI Content Safety
    - The importance of text moderation
    - Steps involved in implementing text moderation
  - Moderating Images with Azure AI Content Safety
    - The importance of image moderation
    - Steps involved in implementing image moderation
  - Detecting Jailbreak Risks
  - Detecting Protected Material
  - Content Filtering for Text Moderation
  - Practical: Implementing a Text Moderation Solution with Azure AI Content Safety
    - Setting up the Azure AI Content Safety resource
    - Configuring environment variables
    - Installing the Azure AI Content Safety Python package
    - Extending the handling of the content
- Chapter Review
- Chapter Quiz
5. Implementing Computer Vision Solutions with Azure AI
- Introduction to Azure AI Vision
  - What Is Computer Vision?
  - What Is Azure AI Vision?
  - Case Study
- Image Analysis with Azure AI Vision
  - The Fundamentals of Image Analysis
  - Performing Image Analysis
    - Selecting the appropriate visual features
    - Detecting objects in images and generating image tags
    - Interpreting image processing responses
  - Extracting Text with Azure AI Vision
    - Choosing the workload
    - Practical: Extracting and converting handwritten text
- Facial Recognition and Analysis
  - Fundamentals of Facial Recognition
    - Considerations
    - Functionalities
    - Response from performing the analysis
  - Practical: Implementing Facial Recognition in Your Application
- Custom Vision and Object Detection
  - Building Custom Image Classification Models
    - Data collection and preparation
    - Choosing the appropriate model
    - Accuracy trade-offs and decision frameworks
    - Performance benchmarks for different types of hardware
    - Model optimization techniques
    - Labeling images
  - Implementing Object Detection
    - Training the custom image model
    - Evaluating the custom vision model metrics
    - Publishing and consuming a custom vision model
  - Practical: Creating a Custom Vision Object Detection Solution
- Working with Video Content
  - Using Azure AI Video Indexer
  - Practical: Analyzing Video Content with Azure AI Video Indexer
- Chapter Review
- Chapter Quiz
6. Implementing Natural Language Processing Solutions
- Fundamentals of Natural Language Processing
  - Introduction to NLP
  - Core Components of NLP
    - Transformer architectures and attention mechanisms
    - Pretrained models
    - Basic NLP code example
  - Common NLP Techniques and Algorithms
    - Deeper implementation details and optimization
    - Handling different languages and writing systems
    - Failure modes and edge cases
    - Large-scale deployment and real-time processing
  - A Look into NLP in Microsoft Azure
- Introduction to the Azure AI Language Service
  - Understanding Azure AI Language
  - Using Prebuilt Solutions
    - Extracting key phrases from text
    - Extracting entities from text
    - Performing sentiment analysis
    - Detecting the language used in text
    - Detecting personally identifiable information
  - Using Azure AI Speech to Process Speech
    - Understanding Azure AI Speech
    - Implementing Prebuilt Speech Solutions
      - Implementing speech-to-text solutions
      - Implementing text-to-speech solutions
      - Implementing intent recognition
      - Implementing keyword recognition
    - Implementing Custom Speech Solutions
- Translating with Azure AI Translator
  - Understanding Azure AI Translator
    - Technical details on neural machine translation and formatting preservation
    - Optimizing batch translation
  - Implementing Prebuilt Translation Solutions
    - Translating text and documents
    - Translating speech to text and speech to speech
  - Implementing Custom Translation Solutions
  - Practical: Building a Custom Translation Solution
- More Best Practices
- Chapter Review
- Chapter Quiz
7. Advanced NLP Techniques and Language Understanding
- Working with Language Understanding Models
  - Creating Intents, Utterances, and Entities
  - Building Language Understanding Models
    - Training the model to recognize intents and entities
    - Evaluating model performance with labeled datasets
    - Handling ambiguous user input and improving accuracy
    - Implementing your trained model on a publicly accessible endpoint
    - Analyzing predictions and refining the model
  - Optimizing Language Understanding Models
  - Backing Up and Recovering Language Understanding Models
  - Practical: Building and Integrating Your Own Language Understanding Model
    - Step 1: Create an Azure AI Language resource
    - Step 2: Set up Language Studio
    - Step 3: Define intents, utterances, and entities
    - Step 4: Train your model
    - Step 5: Evaluate and improve your model
    - Step 6: Deploy your model
    - Step 7: Integrate the model into your application
    - Step 8: Clean up your resources
- Building Question-Answering Solutions
  - Understanding Question-Answering Solutions
    - Key features of question-answering solutions
    - Fundamentals of question answering
      - Components of a QA system
      - Types of QA systems
      - Real-world use cases
    - Monitoring usage and identifying areas for improvement
  - Practical: Building Your Own Question-Answering Solution
    - Step 1: Build the question-answering knowledge base
    - Step 2: Train and test your model
    - Step 3: Deploy your model
    - Step 4: Integrate the model into your application
    - Step 5: Integrate with Azure AI Bot Service (optional)
    - Step 6: Clean up your resources
  - Advanced Capabilities
    - Adding multiturn conversations
    - Practical: Implementing multiturn conversations
      - Step 1: Create a question-answering project
      - Step 2: Add new question-and-answer pairs
      - Step 3: Test multiturn conversations
      - Step 4: Deploy your project
      - Step 5: Integrate with applications
    - Example JSON requests and responses
    - Alternate phrasing
    - Practical: Alternate Phrasing
      - Step 1: Access your project in Language Studio
      - Step 2: Add alternate phrasing to a Q&A pair
      - Step 3: Add synonyms and colloquial terms
      - Step 4: Test the Q&A pair
      - Step 5: Add alternate phrasing to a Q&A pair
    - Adding chit-chat
      - Step 1: Add chit-chat to your sources
      - Step 2: Edit chit-chat Q&A pairs
      - Step 3: Add custom chit-chat Q&A pairs
      - Step 4: Test chit-chat integration
      - Step 5: Deploy your updated knowledge base
      - Meeting user expectations and avoiding bias
      - Best practices for chit-chat integration
    - Exporting a knowledge base
- Chapter Review
- Chapter Quiz
8. Implementing Knowledge Mining and Document Intelligence Solutions
- Planning and Implementing a Knowledge-Mining Solution with Azure AI Search
  - Understanding Azure AI Search
    - Components of search
    - The indexing process
  - Creating a Search Service
    - Creating data sources
    - Implementing custom skills
  - Optimizing Search Performance
    - Creating an index
    - Running an indexer
    - Querying an index
    - Working with custom classes
  - Maintaining a Search Solution
    - Security
    - Performance
    - Managing costs
    - Debugging issues
  - Advanced Search Features
    - Using term boosting
    - Scoring profiles
    - Analyzers and tokenized terms
    - Using multiple languages
    - Using semantic search features
    - Using vector search
- Working with Document Intelligence Solutions in Azure AI Document Intelligence
  - Understanding Azure AI Document Intelligence
    - Practical: Setting up a Document Intelligence resource
    - Types of models
    - Using a custom model
    - Practical: Using prebuilt models
      - Integrating with Azure AI Search
      - Indexing data from Azure Blob Storage
  - Practical: Implementing a Custom Document Intelligence Model
    - Step 1: Prepare your training data
    - Step 2: Upload your training data to Azure Blob Storage
    - Step 3: Train your custom model
    - Step 4: Test your custom model
    - Step 5. Use the custom model in your application
  - Practical: Creating a Composed Document Intelligence Model
    - Step 1: Train separate models for each document type
    - Step 2: Create a composed model
    - Step 3. Deploy and use your composed model
  - Practical: Building Custom Skills for Azure AI Search
    - Defining a custom skill in Azure AI Search
    - Implementing a custom function
    - Creating an index and an indexer
    - Running the indexer
    - Querying the processed content
- Chapter Review
- Chapter Quiz
9. Utilizing the Azure OpenAI Service for Generative AI Applications
- Generative AI on Microsoft Azure
  - Types of Generative AI Models
    - OpenAI o-series models
      - Capabilities
      - Use cases
    - GPT-4 models
      - Capabilities
      - Use cases
    - Embedding models
      - Capabilities
      - Use cases
    - DALL-E models
      - Capabilities
      - Use cases
    - Choosing a model
    - Limitations and ethical considerations
  - Building Responsible Generative AI Solutions
    - Identifying and mitigating potential harms
    - Measuring potential harms
    - Mitigating potential harms
    - Working with responsible generative AI solutions
- Prompt Engineering with the Azure OpenAI Service
  - Writing Effective Prompts
  - Advanced Techniques and Best Practices
- Generating Content with the Azure OpenAI Service
  - Using the Azure OpenAI Service
    - The completions playground
      - Temperature
      - Max length (in tokens)
      - Stop sequences
      - Top P (aka nucleus sampling)
      - Frequency penalty
      - Presence penalty
      - Pre-response text
      - Post-response text
    - The chat playground
  - Using Azure OpenAI in Your Applications
  - Generating Text
    - Practical: Using Azure OpenAI API GPT-4 to generate text
      - Step 1: Set up an Azure OpenAI Service resource
      - Step 2: Set up your development environment
      - Step 3: Implement the GPT-4 text generation script
      - Step 4: Run the script
    - Best practices
  - Generating Images
    - Practical: Generating images with the Azure OpenAI DALL-E API
      - Step 1: Deploy the DALL-E model
      - Step 2: Update environment variables
      - Step 3: Implement a DALL-E image generation script
      - Step 4: Run the script
    - Best practices
  - Generating Code
    - Fixing a bug in a piece of code
    - Writing unit tests
    - Building complete functions and classes
    - Automated code documentation
    - Practical: Creating a chatbot
      - Step 1: Deploy the GPT-3.5 Turbo model
      - Step 2: Install the required Python packages
      - Step 3: Update the environment variables
      - Step 4: Implement the chatbot script
      - Step 5: Run the chatbot
    - Best practices
- Fine-Tuning and Optimizing Generative AI Models
  - Fine-Tuning Your OpenAI Model
  - Integrating Data Sources
  - Interacting with the Model
- Retrieval-Augmented Generation
  - Understanding RAG
    - Real-life use cases of RAG
    - Using RAG in Azure OpenAI
    - Performance optimization strategies for RAG
  - Practical: Implementing RAG with Azure OpenAI and Azure AI Search
    - Step 1: Create an Azure AI Search resource
    - Step 2: Deploy models in Azure OpenAI
    - Step 3: Set up an Azure AI Search index
    - Step 4: Configure environment variables
    - Step 5: Implement the RAG solution
    - Step 6: Run the script
    - Step 7: Test and refine the application
    - Best practices
- Chapter Review
- Chapter Quiz
10. The Future of AI in Microsoft Azure
- A Look at Key Trends
  - Advances in Complex Reasoning
    - What makes these models different?
    - Azures playground for reasoning engines
      - Why this matters now
    - Small Language Models
      - Technologies and tools in Azure
      - Integration of SLMs with Azure Services
      - Building with SLMs in Azure
      - Implications for Azure
    - Multimodal AI
      - Technologies and tools in Azure
      - Applications and implications
    - Agentic AI
    - The Integration of Azure AI with the Entire Azure Platform
      - AI-driven analytics and insights
      - AI-enhanced security
      - Integrating AI with the development environment
      - Holistic AI architecture
      - A comprehensive ecosystem
    - Practical: Creating a Complex Architecture on Azure with AI Services
      - Step 1: Deploy Azure Storage for data ingestion
      - Step 2: Configure Azure AI Services for data processing
      - Step 3: Set up Azure AI Search
      - Step 4: Deploy Azure Machine Learning
      - Step 5: Integrate the Azure OpenAI Service
      - Step 6: Build and deploy AI models
      - Step 7: Integrate and secure the architecture
      - Step 8: Create and deploy a web application for user interaction
  - Streamlining the AI Development Process
  - Developing with Microsoft Copilot
    - Streamlining the development process with Azure OpenAI and the Teams AI library
    - Enhancing productivity with AI and Copilot in Microsoft Power Platform
  - Practical: Using Copilot in the Azure Ecosystem (with Code)
    - Step 1: Install and launch Azure Data Studio
    - Step 2: Set up GitHub Copilot
    - Step 3: Sign in to GitHub
    - Step 4: Use GitHub Copilot in Azure Data Studio to write SQL queries with Copilot assistance
  - Working with Prompt Flow in Azure AI Foundry
    - Phase 1: Initialization
    - Phase 2: Experimentation
    - Phase 3: Evaluation
    - Phase 4: Refinement
    - Phase 5: Production
- Microsoft Fabric
  - Building AI Applications with Microsoft Fabric
  - Fostering a Data Culture Across Organizations
    - Ways to support the adoption of a data culture
    - Obstacles and solutions for the development of a data culture
  - Practical: Creating an End-to-End Solution with Microsoft Fabric
    - Step 1: Set up your environment
    - Step 2: Create a lakehouse
    - Step 3: Ingest data
    - Step 4: Transform data
    - Step 5: Create reports
    - Step 6: Scheduling and orchestration (optional)
    - Step 7: Clean up
- A Closing Note
- Chapter Quiz
A. Answer Keys
- Chapter 1 Answer Key
- Chapter 2 Answer Key
- Chapter 3 Answer Key
- Chapter 4 Answer Key
- Chapter 5 Answer Key
- Chapter 6 Answer Key
- Chapter 7 Answer Key
- Chapter 8 Answer Key
- Chapter 9 Answer Key
- Chapter 10 Answer Key
Index