Microsoft Azure AI Engineer Associate

Microsoft · AI-102 · Associate

Microsoft · Microsoft Azure + M365

Microsoft Azure AI Engineer Associate

AI-102● retiringAssociate
Official Microsoft source · learn.microsoft.com ↗

AI-102 · ● Retiring (June 30, 2026) · Associate · Microsoft

Critical notice: This exam retires on June 30, 2026 at 11:59 PM Central Standard Time. The replacement exam is AI-103 (Azure AI App and Agent Developer Associate), which launches June 2026 and shifts focus from Azure AI service integration toward generative AI applications and autonomous agent development. AI-102 holders will retain their credential indefinitely; the exam retires but the certification does not expire.

Exam facts

FieldValue
Cost$165 USD; regional pricing may vary
Duration130 minutes
Questions40–60 mixed format (multiple choice, scenario, drag-and-drop, labs)
Passing700/1000 scaled score
FormatMultiple choice, multiple response, drag-and-drop, hands-on labs, scenario-based
DeliveryPearson VUE (test center and OnVUE remote proctoring)
LanguagesEnglish; regional language options available
Valid3 years from pass date
RenewalRenewal exam (AI-102) or higher Microsoft AI cert (e.g., AI-103, AI-200 series) to extend credential
PrerequisitesNone; C#, Python, or JavaScript proficiency recommended; Azure fundamentals (AI-900) helpful but not required
ReleasedJanuary 2021
RetiringJune 30, 2026 11:59 PM CST

Vendor source — Microsoft Certified: Azure AI Engineer Associate ↗

Official exam guide — Exam AI-102: Designing and Implementing a Microsoft Azure AI Solution ↗

Exam study guide — Study Guide for Exam AI-102 ↗

About

The Microsoft Azure AI Engineer Associate certification (AI-102) validates intermediate-level expertise in designing, building, and deploying AI solutions on Microsoft Azure. Launched in January 2021, it became the standard credential for software developers, cloud engineers, and AI specialists integrating Azure Cognitive Services (now Azure AI Services), Azure Machine Learning, Azure OpenAI, and knowledge mining tools into production applications. The exam covers six domains: planning/managing Azure AI solutions, implementing decision support, computer vision, natural language processing, knowledge mining, and generative AI. As of May 2026, Microsoft is retiring AI-102 in favor of AI-103, which emphasizes generative AI applications and autonomous agent development using Azure AI Foundry—a strategic shift reflecting the market move toward agentic systems. Candidates have until June 30, 2026 to earn AI-102; those holding the certification retain it indefinitely.

Domain context — Data/AI

Intermediate applied artificial intelligence and machine learning services spanning hyperscale public cloud platforms. This domain covers designing and implementing end-to-end AI solutions using managed Azure AI services, integrating vision, language, and generative AI workloads into production applications, and applying responsible AI governance at scale.

Read full deep dive — Microsoft Azure + M365 Ecosystem →

Topics covered

The AI-102 exam blueprint covers six domains with the following weight distribution:

  • Plan and Manage an Azure AI Solution (15–20%) — Select appropriate Azure AI services for solution requirements; plan resource deployment; implement content moderation and safety controls; manage authentication, costs, and monitoring; apply responsible AI principles.

  • Implement Decision Support Solutions (10–15%) — Build machine learning pipelines; train and evaluate models; use Azure Machine Learning for regression, classification, and clustering; deploy models as web services; implement AutoML for rapid prototyping.

  • Implement Azure AI Vision Solutions (15–20%) — Use Azure Computer Vision API for image classification, object detection, optical character recognition; deploy Form Recognizer for document intelligence; implement facial recognition; apply ethical constraints.

  • Implement Natural Language Processing Solutions (15–20%) — Build solutions with Azure Language Service for sentiment analysis, entity extraction, key phrase recognition; implement Azure Translator for multilingual support; deploy named entity recognition (NER); handle language-specific edge cases.

  • Implement Knowledge Mining Solutions (10–15%) — Extract structured data from unstructured sources using Azure AI Search; implement indexing and retrieval pipelines; use Document Intelligence for form and table extraction; build knowledge bases and Q&A systems.

  • Implement Generative AI Solutions (10–15%) — Integrate Azure OpenAI Service (GPT models); implement prompt engineering and prompt flows; design retrieval-augmented generation (RAG) architectures; apply safety guardrails; optimize costs and latency.

Source: Official Exam Objectives ↗

Common skills at Data/AI · Associate

Shared content for the Data/AI domain at Associate level — not specific to this cert.

  • Design end-to-end ML/AI pipeline architecture
  • Evaluate model performance: precision, recall, F1, ROC-AUC, RMSE
  • Diagnose and mitigate bias and drift in production models
  • Implement monitoring, alerting, and retraining workflows
  • Apply responsible AI frameworks: fairness, transparency, privacy, accountability
  • Integrate third-party APIs and SDKs into production code
  • Optimize model serving: latency, throughput, cost
  • Document AI decisions for regulatory compliance (GDPR, HIPAA, etc.)

Recommended courses at Data/AI · Associate

ProviderTitleCostURL
Microsoft Learn (Official)AI-102T00-A: Develop AI solutions in AzureFree↗
Microsoft Learn (Official)Exam Readiness Zone: AI-102 (6-part series)Free↗
YouTube (Free, by John Savill)AI-102 Study Cram: Azure AI Engineer Associate CertificationFree↗
Coursera (Official Partner)Exam Prep AI-102: Microsoft Azure AI Engineer AssociateFree audit; $39+ for certificate↗
PluralsightAzure AI Engineer Associate (AI-102): Azure AI Fundamentals, Planning, and Management$399/year (or $29/month)↗
UdemyAI-102: Microsoft Azure AI Engineer Associate Complete Exam Prep$12–$15 (during sales)↗
KodeKloudAI-102: Microsoft Certified Azure AI Engineer Associate Course$50/month↗

Course-selection rule: Microsoft Learn's official AI-102T00-A course and Exam Readiness videos align precisely with exam objectives and include hands-on labs with free Azure trial credits. John Savill's free YouTube cram covers the big picture efficiently. Coursera provides structured, scaffolded pacing suitable for career-changers. Pluralsight and Udemy offer alternative depth suitable for self-directed learners.

Practice exams

ProviderTitleCostURL
MeasureUp (Microsoft Official Partner)AI-102 Practice Test$99↗
WhizlabsExam AI-102: Azure AI Engineer Associate (15+ practice questions)$49–$99↗
Tutorials DojoAI-102 Practice Exams (Full Bundle)$15–$25↗
Microsoft LearnFree Practice Test (limited; via exam dashboard)Free↗

Books

TitleAuthorPublisherYearISBNURL
Azure AI Engineer Associate (AI-102) Study Guide: In-Depth Certification Guide and PracticeRenaldi GondosubrotoO'Reilly2024978-1-0981-6926-8↗
Azure AI-102 Certification Essentials: Strengthen your Azure AI foundation and ace the AI-102 exam with real-world cases and mock testsPeter T. LeePackt2024978-1-8362-0527-2↗
AI-102: Designing and Implementing a Microsoft Azure AI Solution Study Guide with Practice Questions and LabsNouman Ahmed KhanIPSpecialist2023979-8-5280-9062-7↗
Exam AI-102: Designing and Implementing a Microsoft Azure AI SolutionClark HamiltonSelf-published2024979-8-3263-0157-4↗

Book rule: All titles reference AI-102 specifically. Renaldi Gondosubroto's O'Reilly guide (2024) is the most comprehensive and current; Peter Lee's Packt edition emphasizes real-world case studies and is suitable for practitioners integrating AI into existing systems.

Typical job titles at Data/AI · Associate

Azure AI Engineer · Applied AI Engineer · Cloud Developer (AI focus) · Solutions Architect (AI/ML) · ML Operations Engineer · AI Solutions Specialist · Conversational AI Developer · Computer Vision Specialist · Azure Data/AI Engineer

(Job titles drawn from current job-board postings that list AI-102 or Azure AI Engineer expertise as required or strongly preferred.)

Salary

RegionRangeSource
USD$120,000 – $200,000+Glassdoor ↗ · ZipRecruiter ↗ · Levels.fyi ↗
GBP£57,600 – £92,000IT Jobs Watch ↗ (UK AI/ML mid-level roles) · Hays ↗
ZARNo region-specific data available — use Azure Solutions Architect as proxy (R600,000–R1,200,000 annually for mid-level)CareerJunction ↗ · PayScale ZA ↗

Salary rule: AI-102 holders typically earn 20–30% above entry-level data science roles (AI-900) and 10–15% below senior architect or ML director roles. ZAR data unavailable; South African job boards should be checked directly for Azure AI Engineer postings, as limited AI-102-specific hiring data exists in the region. Actual compensation varies significantly by company size, location (Bay Area/London premium +25–40%), and equity in tech-sector roles.

Skills validated

Cert-specific — what this exam actually tests, distinct from the shared "Common skills" above.

  • Integrate Azure Cognitive Services (Computer Vision, Language Service, Translator) into applications
  • Design and deploy Azure Machine Learning pipelines for classification and regression
  • Implement content moderation and safety controls (content filtering, bias detection)
  • Build Azure AI Search solutions for knowledge mining and semantic retrieval
  • Deploy and manage Azure OpenAI Service endpoints with cost optimization
  • Implement prompt engineering and retrieval-augmented generation (RAG) patterns
  • Develop computer vision applications: image classification, object detection, OCR, form recognition
  • Build NLP applications: sentiment analysis, named entity recognition, multilingual translation
  • Implement responsible AI governance: ethical guardrails, monitoring, fairness audits
  • Use Azure SDK (Python, C#, JavaScript) to consume Azure AI services from production code
  • Optimize ML model performance: hyperparameter tuning, cross-validation, ensemble methods
  • Deploy models as containerized or serverless endpoints (Azure Container Instances, App Service, Functions)

Related certifications

Sources

Last verified: 2026-05-01

Parent ecosystem: Microsoft Azure + M365 Ecosystem

Parent domain: Data/AI Domain

Vendor overview: Microsoft Vendor Overview

Roles that use this certification

1 career-path guide on this site put this exam in the sequence.

Data & AIAI Engineer
Rate this cert
…
Was this helpful?
Comments (—)
0/2000