🚀 Launch Special: $29/mo for life --d --h --m --s Claim Your Price →
AI-102
AI-102 Microsoft Azure

AI Engineer

The AI-102 certification course teaches Azure AI Engineers to design, build, and deploy production-ready AI solutions using Azure Cognitive Services, Machine Learning, and Knowledge Mining, covering decision support, computer vision, NLP, and document intelligence.

120
Minutes
50
Questions
700/1000
Passing Score
$165
Exam Cost
7
Languages

Who Should Take This

It is intended for software developers, data engineers, or solution architects who have at least one year of hands‑on Azure experience and a solid grasp of AI fundamentals (AI‑900). These professionals aim to validate their ability to implement end‑to‑end AI services on Azure and advance to roles that design and manage enterprise‑scale AI solutions.

What's Covered

1 Planning Azure AI resources, managing AI service accounts, configuring diagnostic logging, and implementing responsible AI practices.
2 Implementing Azure AI Content Safety for text and image moderation, configuring content filters, and managing moderation policies.
3 Implementing image analysis, custom vision models, face detection, optical character recognition, and video analysis with Azure AI Vision.
4 Implementing text analytics, language understanding, question answering, conversational language understanding, and custom text classification.
5 Implementing Azure AI Search, document intelligence for form processing, and knowledge mining pipelines with cognitive skills.
6 Implementing Azure OpenAI Service, prompt engineering, retrieval augmented generation, and responsible generative AI deployment.

Exam Structure

Question Types

  • Multiple Choice
  • Multiple Response
  • Case Studies

Scoring Method

Scaled score 100-1000, passing score 700

Delivery Method

Proctored exam, 40-60 questions, 100 minutes

Prerequisites

None required. AI-900 recommended.

Recertification

Renew annually via free Microsoft Learn renewal assessment

What's Included in AccelaStudy® AI

Adaptive Knowledge Graph
Practice Questions
Lesson Modules
Console Simulator Labs
Exam Tips & Strategy
20 Activity Formats

Course Outline

79 learning goals
1 Domain 1: Plan and Manage an Azure AI Solution
4 topics

Select the appropriate Azure AI service

  • Identify Azure AI Services offerings and explain the capabilities of multi-service resources, single-service resources, and the distinction between prebuilt, customizable, and custom AI models.
  • Identify Azure Applied AI services including Azure Bot Framework, Azure AI Search, Azure AI Document Intelligence, and Azure OpenAI Service and explain how each addresses specific AI workload patterns.
  • Select the appropriate Azure AI service based on solution requirements including latency, throughput, data residency, customization needs, and cost constraints for vision, language, speech, and decision workloads.
  • Analyze tradeoffs between using Azure AI Services prebuilt models, customizing existing models, and building custom models to determine the optimal approach for a given AI solution scenario.

Plan, create, and configure Azure AI resources

  • Identify Azure AI resource provisioning options and explain the differences between multi-service and single-service deployments, region selection, pricing tiers, and resource group organization.
  • Create and configure Azure AI resources including selecting pricing tiers, configuring networking settings, enabling managed identities, and organizing resources within resource groups for production deployments.
  • Configure diagnostic logging, usage monitoring, alert rules, and container-based deployment of Azure AI Services for operational visibility, edge scenarios, and on-premises deployments with connected and disconnected configurations.

Manage AI service security

  • Identify Azure AI Services authentication mechanisms and explain the differences between key-based authentication, Microsoft Entra ID token-based authentication, and managed identity access patterns.
  • Implement key management and rotation strategies for Azure AI Services using Azure Key Vault, managed identities, and role-based access control to secure API credentials and service access.
  • Implement network security for Azure AI resources using virtual network integration, private endpoints, service endpoints, and firewall rules to restrict access to AI service endpoints.
  • Analyze security architecture for Azure AI solutions and determine the appropriate combination of authentication, network isolation, encryption, and access control mechanisms for compliance requirements.

Plan and implement responsible AI practices

  • Identify the Microsoft Responsible AI principles and explain how fairness, reliability, safety, privacy, security, inclusiveness, transparency, and accountability apply to Azure AI solution design.
  • Implement responsible AI practices in Azure AI solutions including transparency documentation, error analysis, fairness assessment, and human oversight mechanisms for deployed AI models.
  • Analyze AI solution designs for responsible AI compliance and determine appropriate mitigation strategies for bias, safety risks, privacy concerns, and lack of transparency in production deployments.
2 Domain 2: Implement Decision Support Solutions
2 topics

Implement content moderation with Azure AI Content Safety

  • Identify Azure AI Content Safety capabilities and explain how text moderation, image moderation, and severity levels detect harmful content across categories including hate, violence, sexual, and self-harm.
  • Implement text and image content moderation using Azure AI Content Safety API to analyze inputs, interpret severity scores across harm categories, and configure threshold-based filtering for user-generated content.
  • Implement custom blocklists and custom content categories in Azure AI Content Safety to define prohibited terms and train organization-specific moderation classifiers for domain-specific content filtering.
  • Analyze content moderation architectures and determine the optimal combination of built-in categories, custom categories, and blocklists to meet application-specific safety requirements while minimizing false positives.

Implement Personalizer recommendations

  • Identify Azure AI Personalizer capabilities and explain how the rank and reward loop, exploration versus exploitation settings, and feature engineering drive personalized recommendation decisions.
  • Implement a Personalizer recommendation loop by configuring actions, context features, reward signals, and evaluation settings to deliver personalized content ranking in real-time applications.
  • Analyze Personalizer model performance using offline evaluation and reward trends to determine when to adjust exploration rates, retrain models, or modify feature definitions for improved recommendation quality.
3 Domain 3: Implement Computer Vision Solutions
5 topics

Analyze images using Azure AI Vision

  • Identify Azure AI Vision Image Analysis capabilities and explain how image tagging, object detection, caption generation, dense captioning, smart cropping, and people detection features process visual content.
  • Implement image analysis using Azure AI Vision to extract tags, detect objects with bounding boxes, generate captions, and retrieve image metadata for content understanding workflows.
  • Implement image embeddings for multi-modal vector search and spatial analysis using Azure AI Vision to enable similarity queries, detect people presence in video streams, and process occupancy events.

Implement video analysis

  • Identify Azure Video Indexer capabilities and explain how video and audio insights including transcription, face detection, scene segmentation, topic extraction, and sentiment analysis are derived from media content.
  • Implement video analysis using Azure Video Indexer to upload and index video content, extract insights including transcripts and detected faces, and customize language and brand models for domain-specific accuracy.

Implement image classification and object detection

  • Identify Azure AI Custom Vision capabilities and explain how custom image classification and object detection models are trained, evaluated, published, and consumed through prediction endpoints.
  • Implement custom image classification models using Azure AI Custom Vision including project creation, training image upload, domain selection, model training, performance evaluation, and endpoint publishing.
  • Implement custom object detection models using Azure AI Custom Vision including tagged region annotation, model training with domain-specific data, iteration management, and prediction API integration.
  • Analyze custom vision model performance using precision, recall, and mAP metrics to determine when to add training data, adjust probability thresholds, or switch between compact and standard domain models.

Read and process text with OCR and Document Intelligence

  • Identify Azure AI Vision OCR capabilities and explain how the Read API extracts printed and handwritten text from images and documents with line-level and word-level bounding box results.
  • Implement text extraction from images and documents using the Azure AI Vision Read API including handling multi-page documents, interpreting confidence scores, and processing structured text output.
  • Analyze OCR results quality and determine when to use the Vision Read API versus Azure AI Document Intelligence based on document structure complexity, extraction accuracy needs, and field-level parsing requirements.

Implement facial detection and analysis

  • Identify Azure AI Face service capabilities and explain how face detection, face attributes, face verification, face grouping, and face identification work within responsible AI usage constraints.
  • Implement face detection and verification workflows using the Azure AI Face service including detecting faces, extracting attributes, and configuring person groups with the Limited Access approval process.
  • Analyze face detection solution designs for responsible AI compliance including Limited Access requirements, bias mitigation, privacy considerations, and appropriate use-case boundaries for facial recognition technology.
4 Domain 4: Implement Natural Language Processing Solutions
6 topics

Analyze text using Azure AI Language

  • Identify Azure AI Language text analytics capabilities and explain how language detection, sentiment analysis, key phrase extraction, entity recognition, entity linking, and PII detection process unstructured text.
  • Implement text analytics using Azure AI Language to perform sentiment analysis, extract key phrases, recognize and link entities, detect PII, and generate extractive and abstractive summaries for document processing workflows.
  • Analyze text analytics results across multiple language features and determine the optimal combination of entity recognition, sentiment, key phrases, PII detection, and summarization for a given NLP solution scenario.

Build question answering solutions

  • Identify Azure AI Language custom question answering capabilities and explain how knowledge bases, QnA pairs, metadata, multi-turn conversations, and active learning improve answer accuracy.
  • Implement a custom question answering knowledge base using Azure AI Language including creating QnA pairs from documents and URLs, configuring alternate questions, and deploying the knowledge base for inference.
  • Implement multi-turn conversation flows and active learning in custom question answering to handle follow-up questions, refine answer quality, and manage knowledge base content lifecycle.
  • Analyze question answering solution performance using confidence scores, test results, and user feedback to determine when to add QnA pairs, adjust thresholds, or restructure the knowledge base.

Build conversational language understanding

  • Identify conversational language understanding (CLU) concepts and explain how intents, entities, utterances, and prebuilt components model conversational interactions in Azure AI Language.
  • Implement a conversational language understanding project using Azure AI Language including defining intents, labeling entities in utterances, training the model, and deploying to a prediction endpoint.
  • Implement orchestration workflows that combine CLU with custom question answering and other Azure AI Language features to create unified conversational AI applications with active learning feedback loops.
  • Analyze CLU model performance using evaluation metrics and confusion matrices to identify intent misclassification patterns, entity extraction gaps, and training data improvements needed for production accuracy.

Implement custom text classification and named entity recognition

  • Identify custom text classification and custom NER capabilities in Azure AI Language and explain how single-label, multi-label classification, and custom entity extraction projects are structured.
  • Implement custom text classification models using Azure AI Language including data labeling, single-label and multi-label project configuration, model training, evaluation, and deployment for inference.
  • Implement custom named entity recognition models using Azure AI Language including entity type definition, document annotation, model training, and extraction pipeline integration for domain-specific entities.
  • Analyze custom classification and NER model performance and determine strategies for improving precision, recall, and F1 scores through data quality improvements, annotation consistency, and model iteration.

Implement Azure AI Speech solutions

  • Identify Azure AI Speech service capabilities and explain how speech-to-text, text-to-speech, speech translation, speaker recognition, intent recognition, and keyword recognition enable voice-based AI applications.
  • Implement speech-to-text transcription using Azure AI Speech including real-time and batch transcription, language configuration, and custom speech model training for domain-specific vocabulary recognition.
  • Implement text-to-speech synthesis and speech translation using Azure AI Speech including voice selection, SSML prosody control, custom neural voice creation, and real-time speech-to-speech translation across languages.
  • Implement intent recognition using Azure AI Speech integrated with conversational language understanding to extract user intent and entities directly from spoken audio input in voice-driven applications.
  • Analyze speech solution requirements and determine the optimal combination of speech-to-text, text-to-speech, translation, and custom model training for production accuracy and latency targets.

Translate language with Azure AI Translator

  • Identify Azure AI Translator capabilities and explain how text translation, document translation, transliteration, language detection, and dictionary lookup handle multi-language content processing.
  • Implement text and document translation using Azure AI Translator including standard translation, transliteration, batch document translation, and Custom Translator training with parallel corpora for domain-specific accuracy.
  • Analyze translation quality and determine when to use standard translation, custom translation models, or speech translation based on domain terminology accuracy, document format preservation, and real-time requirements.
5 Domain 5: Implement Knowledge Mining and Document Intelligence Solutions
2 topics

Implement Azure AI Search solutions

  • Identify Azure AI Search components and explain how search indexes, indexers, data sources, skillsets, knowledge stores, and scoring profiles compose a knowledge mining pipeline.
  • Implement Azure AI Search indexes and indexers including field definitions, analyzers, scoring profiles, data source connections, field mappings, change detection, and scheduling for automated content ingestion.
  • Implement AI enrichment skillsets in Azure AI Search using built-in cognitive skills, custom Web API skills, and Azure AI Services integration to extract entities, key phrases, and structured data during indexing.
  • Implement knowledge store projections in Azure AI Search to persist enriched content as tables, objects, and files in Azure Storage for downstream analytics and reporting scenarios.
  • Analyze Azure AI Search solution performance and determine optimization strategies for relevance tuning, scoring profile adjustments, skillset configuration, and index schema refinement.

Implement Azure AI Document Intelligence solutions

  • Identify Azure AI Document Intelligence capabilities and explain how prebuilt models for invoices, receipts, ID documents, business cards, and the general document model extract structured data from documents.
  • Implement document analysis using Azure AI Document Intelligence prebuilt models to extract key-value pairs, tables, and document-specific fields from invoices, receipts, and identity documents.
  • Implement custom document extraction models using Azure AI Document Intelligence including training data preparation, custom model training, composed model creation, and extraction pipeline integration.
  • Analyze document intelligence solution architectures and determine the optimal combination of prebuilt models, custom models, and composed models for multi-document-type processing workflows.
6 Domain 6: Implement Generative AI Solutions
2 topics

Use Azure OpenAI Service

  • Identify Azure OpenAI Service capabilities and explain how GPT, DALL-E, and embedding model families support text generation, code generation, image generation, and vector embedding use cases.
  • Identify prompt engineering techniques and explain how system messages, few-shot examples, chain-of-thought prompting, and parameter tuning control generative AI output quality and behavior.
  • Implement Azure OpenAI Service resource provisioning and model deployment including resource creation, model selection, deployment configuration, quota management, and content filtering policy assignment.
  • Implement prompt engineering strategies for Azure OpenAI including system message design, few-shot example selection, output format specification, and generation parameter optimization for text and code completion tasks.
  • Implement image generation using Azure OpenAI DALL-E and embedding generation using embedding models for vector search, semantic similarity, and downstream retrieval-augmented generation integration.

Optimize generative AI solutions

  • Identify RAG architecture components and explain how Azure AI Search vector search, embedding generation, and Azure OpenAI completions combine to ground generative AI responses in organizational data.
  • Implement a RAG solution using Azure AI Search and Azure OpenAI including document chunking, embedding generation, vector index configuration, hybrid search queries, and context-grounded completion requests.
  • Implement Azure OpenAI model fine-tuning workflows including training data preparation, fine-tuning job configuration, model evaluation, and deployment of fine-tuned models for domain-specific generation tasks.
  • Implement content filtering policies for Azure OpenAI deployments including configuring severity thresholds for hate, violence, sexual, and self-harm categories, and integrating blocklists for custom content control.
  • Analyze RAG solution quality and determine optimization strategies for chunking size, embedding model selection, retrieval relevance, context window management, and grounding accuracy in production deployments.
  • Analyze generative AI deployment architectures and determine the optimal configuration of content filtering, throughput provisioning, model versioning, fine-tuning, and responsible AI compliance for production operation.

Hands-On Labs

20 labs ~371 min total Console Simulator

Practice in a simulated cloud console or Python code sandbox — no account needed. Each lab runs entirely in your browser.

Certification Benefits

Salary Impact

$145,000
Average Salary

Related Job Roles

AI Engineer ML Engineer Cognitive Services Developer AI Solutions Developer

Industry Recognition

Microsoft Azure certifications are among the most valued in enterprise IT, with Microsoft holding the second-largest cloud market share globally and serving as the dominant platform in enterprise and hybrid cloud environments.

Scope

Included Topics

  • All domains and task statements in the Microsoft Azure AI Engineer Associate (AI-102) exam guide: Domain 1 Plan and Manage an Azure AI Solution (15-20%), Domain 2 Implement Decision Support Solutions (10-15%), Domain 3 Implement Computer Vision Solutions (15-20%), Domain 4 Implement Natural Language Processing Solutions (30-35%), Domain 5 Implement Knowledge Mining and Document Intelligence Solutions (10-15%), and Domain 6 Implement Generative AI Solutions (10-15%).
  • Associate-level Azure AI engineering responsibilities including planning AI service selection, configuring security, managing Azure AI resources, implementing responsible AI practices, content moderation, computer vision, natural language processing, knowledge mining, document intelligence, and generative AI solutions.
  • Key Azure AI services: Azure AI Services (multi-service and single-service resources), Azure AI Content Safety, Azure AI Vision (Image Analysis, OCR, Face), Azure AI Custom Vision, Azure Video Indexer, Azure AI Language (text analytics, question answering, conversational language understanding, custom text classification, custom NER), Azure AI Speech (speech-to-text, text-to-speech, speech translation, speaker recognition), Azure AI Translator, Azure AI Search (indexes, indexers, skillsets, knowledge store), Azure AI Document Intelligence (prebuilt and custom models), Azure OpenAI Service (model deployment, prompt engineering, code generation, image generation, RAG).
  • Practical implementation decisions involving Azure AI resource provisioning, networking, authentication, key management, responsible AI principles, monitoring, and integration patterns for production AI solutions on Azure.

Not Covered

  • Advanced machine learning model training, custom neural architecture design, and data science workflows not covered by AI-102 objectives.
  • Azure infrastructure administration, networking, and DevOps topics outside the scope of AI service configuration and deployment.
  • Non-Azure AI platforms, third-party AI tooling, and open-source ML frameworks not integrated with Azure AI Services.
  • Rapidly changing exact service pricing values and temporary commercial offers that are not stable for domain knowledge synthesis.
  • Azure CLI and SDK version-specific API signatures and syntax memorization beyond conceptual service interaction patterns.

Official Exam Page

Learn more at Microsoft Azure

Visit

Ready to master AI-102?

Adaptive learning that maps your knowledge and closes your gaps.

Subscribe to Access

Trademark Notice

Microsoft and Azure are registered trademarks of Microsoft Corporation. Microsoft does not endorse this product.

AccelaStudy® and Renkara® are registered trademarks of Renkara Media Group, Inc. All third-party marks are the property of their respective owners and are used for nominative identification only.