Microsoft Certifications for Data and AI Roles in 2027: What's Current After the 2026 Retirements

Microsoft retired several data and AI certifications in 2025-26, including DP-203, DP-100, AI-102 and AI-900, and introduced AI-103, AI-300 and AI-901. Here are the current certifications for analysts, data engineers, AI engineers and MLOps engineers in 2027, what replaced each retired exam and how renewal works.

GloryTecks Team

Training & Placement Faculty

5 November 2026 6 min read
Share:

If you are planning Microsoft certifications for data and AI roles in 2027, start by checking that the exam still exists. Microsoft retired a large part of its data and AI lineup in 2025-26: the Azure Data Engineer (DP-203), Azure Data Scientist (DP-100), Azure AI Engineer (AI-102) and Azure AI Fundamentals exam AI-900 are gone, along with the Azure Developer and Azure Security Engineer associates. In their place, the current lineup is built around Microsoft Fabric and Microsoft Foundry: DP-700 for data engineers, DP-600 for analytics engineers, PL-300 for Power BI analysts, AI-103 for AI app and agent developers, AI-300 for MLOps engineers and the new AI-901 for AI fundamentals.

Retired exams and what to take instead

Retired certificationRetired onTake instead
Azure Data Engineer Associate (DP-203)2025Fabric Data Engineer Associate (DP-700)
Azure Data Scientist Associate (DP-100)1 June 2026Machine Learning Operations Engineer Associate (AI-300), for production ML
Azure AI Engineer Associate (AI-102)30 June 2026Azure AI Apps and Agents Developer Associate (AI-103)
Azure AI Fundamentals exam AI-90030 June 2026Azure AI Fundamentals exam AI-901
Azure Developer Associate (AZ-204)31 July 2026Role-specific options such as AI-103 for AI applications
Azure Security Engineer Associate (AZ-500)31 August 2026Check Microsoft's current security certifications

The dates for 2026 come from Microsoft's credential retirement page, and the AI-900 date from its study guide, which states that the exam retired on 30 June 2026. Microsoft does not describe these as like-for-like swaps; the newer exams cover newer platforms, so read each study guide rather than assuming the old syllabus carries over.

The current certifications for data and AI roles

CertificationExamFocusDuration
Azure FundamentalsAZ-900Cloud and Azure basics45 minutes
Azure Data FundamentalsDP-900Core data concepts on Azure45 minutes
Azure AI FundamentalsAI-901AI concepts and hands-on Microsoft Foundry basicsSee exam page
Power BI Data Analyst AssociatePL-300Preparing, modelling, visualising and securing Power BI100 minutes
Fabric Analytics Engineer AssociateDP-600Semantic models, lakehouses and warehouses in Fabric100 minutes
Fabric Data Engineer AssociateDP-700Ingesting, transforming and optimising data in Fabric with SQL, PySpark and KQL100 minutes
Azure AI Apps and Agents Developer AssociateAI-103Generative AI and agents with Python and Microsoft Foundry120 minutes
Machine Learning Operations Engineer AssociateAI-300MLOps on Azure Machine Learning and GenAIOps on Foundry120 minutes
Azure Administrator AssociateAZ-104Managing Azure resources, identity and networking100 minutes

The new AI-901 fundamentals exam

AI-901 is more practical than the exam it replaced. Its study guide splits the exam into identifying AI concepts and capabilities (40-45%) and implementing AI solutions by using Microsoft Foundry (55-60%). The hands-on part includes writing system and user prompts, deploying a model in the Foundry portal, building a lightweight chat client with the Foundry SDK, and creating and testing a single-agent solution. For a fundamentals exam, that is a lot of real building, which makes it a useful first step for anyone heading into AI work on Azure.

Paths by role

RoleSuggested path
Data analyst or BI developerDP-900 (optional), then PL-300, then DP-600
Analytics engineerPL-300, then DP-600
Data engineerDP-900 (optional), then DP-700
AI application or agent developerAI-901, then AI-103
MLOps or AI platform engineerAI-901 (optional), then AI-300
Cloud administratorAZ-900, then AZ-104

Fundamentals exams are optional for people with hands-on experience; the associate exams are what employers ask about.

How renewal works

Microsoft role-based certifications expire unless renewed, and renewal is free: you pass an online assessment on Microsoft Learn before the expiry date. PL-300 and DP-600, for example, list a 12-month renewal frequency. Two rules about retired certifications matter: a certification you earned before retirement stays on your transcript until it expires, but once a certification is retired its renewal assessment is no longer available, so it cannot be extended.

If you hold a retired certification

A retired certification is still yours until it expires, so keep it on your resume and LinkedIn with its earned date. What changes is how you present your skills: add the platform you are using now, such as Fabric or Foundry, and, if your role needs a current credential, plan the newer exam that matches your work. For example, a DP-203 holder working on Fabric pipelines would aim for DP-700, and an AI-102 holder building agents would aim for AI-103. Interviewers care far more about recent hands-on work than about which exam code you passed.

Applied Skills changed too

Microsoft Applied Skills, shorter scenario-based credentials, saw many retirements in 2026, including ones for Fabric lakehouses and data warehouses, Azure AI Search knowledge mining, Copilot Studio agents and agentic workflows in Microsoft Foundry. If you planned to use an Applied Skill as a stepping stone, check the current list before you start.

Practical tips for 2027

  1. Check the study guide date before buying any course or practice test; content written before mid-2026 may target retired exams.
  2. Use Microsoft's free practice assessments, now offered through AI Skills Navigator for many exams.
  3. Build while you study. The newer exams, such as AI-103, AI-300 and AI-901, assume you have used Foundry and Fabric, not just read about them.
  4. Plan renewals in your calendar as soon as you pass.

For Power BI specifically, our PL-300 preparation guide covers the exam in detail, and the Microsoft Fabric vs Azure Data Factory comparison explains the platform shift behind DP-700.

Preparing at GloryTecks

The Azure AI Engineer course covers Microsoft Foundry, Azure OpenAI, AI Search, vision, language, speech and document intelligence, useful for AI-901 and AI-103. The Microsoft Fabric Data Engineering course covers OneLake, lakehouses, pipelines, notebooks and real-time intelligence for DP-700, and the Power BI with SQL course covers the skills behind PL-300. All run in classroom batches in Ameerpet and live online; a counsellor can confirm how each batch maps to the current exam guide.

Which Microsoft certification should I take for data engineering in 2027?

The Fabric Data Engineer Associate, exam DP-700, which tests SQL, PySpark and KQL across implementing, ingesting, transforming, monitoring and optimising analytics solutions in Microsoft Fabric. The older Azure Data Engineer Associate (DP-203) is retired.

What replaced AI-102?

The Azure AI Engineer Associate (AI-102) retired on 30 June 2026. The current certification for building AI applications on Azure is the Azure AI Apps and Agents Developer Associate, exam AI-103, built around Python and Microsoft Foundry, with a strong focus on generative AI and agents.

Is AI-900 still available?

No. AI-900 retired on 30 June 2026. Azure AI Fundamentals now uses exam AI-901, which covers AI concepts and capabilities (40-45%) and implementing AI solutions with Microsoft Foundry (55-60%), including deploying models and building a simple agent.

Do Microsoft certifications expire?

Role-based and specialty certifications expire unless renewed. Renewal is free through an online assessment on Microsoft Learn, often every 12 months. If a certification is retired, it stays on your transcript until it expires but can no longer be renewed.

Which Microsoft certification is the most useful for AI careers in 2027?

It depends on the role. AI-103 suits developers building AI applications and agents; AI-300 suits engineers who deploy and operate machine learning and generative AI systems; AI-901 is the starting point for both. Pair any of them with hands-on projects in Microsoft Foundry.

GloryTecks Team

Training & Placement Faculty

The GloryTecks faculty team mentors learners across Data Science, AI and Data Engineering, with a focus on real-time projects and placement readiness in Hyderabad.

Want to master Azure Data Factory?

Learn Data Engineering at GloryTecks in Hyderabad, with trainers who have industry experience, real-time projects and placement assistance.

Explore course