Microsoft Azure AI Apps and Agents Developer Associate
225 practice questions
Last reviewed: April 2026
Personal notes and resource links for your study journey
Filter by Certification
The Microsoft Certified: Azure AI Apps and Agents Developer Associate (Exam AI-103: Developing AI Apps and Agents on Azure) replaces AI-102 and reframes the Azure AI engineer role around building generative apps and agents with Microsoft Foundry. It is a scenario-driven associate exam for developers who plan, build, deploy, and operate AI solutions: choosing models and Foundry services, implementing RAG and multi-agent orchestration, wiring up function-calling and conversation memory, and securing and monitoring everything in production. Coverage spans five areas — planning and managing solutions, generative AI and agentic solutions, computer vision, text analysis, and information extraction — with the heaviest weight on the generative-and-agentic core. Expect Python-oriented questions grounded in the Foundry SDK, Azure AI Search, Azure Content Understanding, Azure Speech, and Azure Translator.
About 25–30%. Choose the right Foundry models and services for the task (LLMs, small language models, multimodal models, Foundry Tools, grounding, vector search, agent workflows), design infrastructure, configure model and agent deployments, and integrate CI/CD. Also covers operations and governance: quotas, scaling, rate limits and cost; monitoring performance, drift, safety and grounding quality; security with managed identity, private networking, keyless credentials and role policies; and responsible AI through safety filters, evaluators, trace logging, and agent oversight controls.
The core, at ~30–35%. Build generative applications (deploy and consume LLM/SLM/code/multimodal models, implement RAG, design tool-augmented and multistep reasoning workflows, evaluate for fabrications/relevance/quality/safety) and build agents (define roles, goals and tool schemas; combine retrieval, function-calling and conversation memory; orchestrate multi-agent solutions; add safeguards and approval flows). Also covers optimization and operations: prompt engineering, reflection and self-critique loops, and observability via tracing, token analytics, safety signals, and latency breakdowns.
About 10–15%. Generate and edit images and video from prompts (including inpainting and mask-based edits), build multimodal understanding workflows (captions, visual question-answering, accessibility alt-text, object/region detection, and Azure Content Understanding in single-task and pro modes), and apply responsible AI for visual content — unsafe-content filters, indirect prompt-injection detection from text embedded in images, and visual policy rules such as watermarks and brand checks.
About 10–15%. Use language models for entity/topic/summary extraction and structured JSON output, detect sentiment, tone and sensitive content, and translate with Azure Translator or LLM-powered flows. The speech half covers speech-to-text and text-to-speech for agentic interactions, speech as an agent modality (including custom speech models), multimodal reasoning from audio, and speech translation.
About 10–15%. Build retrieval and grounding pipelines: ingest and index documents, images, audio and video; configure semantic, hybrid and vector search; enrich content with built-in or custom skills; and run RAG ingestion with OCR. Extract structured content from documents using multimodal OCR + layout + field-extraction pipelines and Azure Content Understanding analyzers that emit clean, grounded, markdown or JSON output for downstream agents.
$110k–$155k–$215k USD annual
Generative-AI and agent-development skills command a premium in 2026. The certification is a credibility signal; the salary range reflects applied AI engineers shipping production LLM and agent systems. Pairing AI-103 with demonstrated Foundry, RAG, and multi-agent project work moves candidates toward the high end. Markets outside major US tech hubs trend lower.
Source: levels.fyi 2025 AI/ML engineer roles, U.S. BLS OEWS May 2024 (15-1252 software developers, 15-2051 data scientists), Glassdoor 2025. Figures are approximate; actual compensation depends on role, region, and experience.
Agent and generative-app development is one of the fastest-growing engineering specialties heading into 2026, and AI-103 is positioned to be the headline Azure credential for it — Microsoft built the exam directly around Foundry, agents, and RAG, the patterns enterprises are racing to ship. Demand is strongest where teams are operationalizing LLMs: grounding on private data, orchestrating multi-agent workflows, and meeting security and responsible-AI requirements. Because it replaces the widely held AI-102, expect it to inherit that exam's recruiter recognition while signaling current, agent-era skills.
There are no formal prerequisites, but AI-103 is a genuine associate-level developer exam. You should be comfortable building apps in Python and familiar with general AI, generative AI, and core Azure services. In practice, candidates do well with prior exposure to the Azure AI services (Language, Vision, Speech, Search, Azure OpenAI) and hands-on time in the Microsoft Foundry portal and SDK.
Microsoft's free Microsoft Learn paths and the AI-103T00 instructor-led course map directly to the five domains. If you previously passed AI-102, budget time to learn what is new rather than re-learning fundamentals: the agent-centric model, multi-agent orchestration, Azure Content Understanding, and the Foundry SDK and tooling are the material that changed. Coming from AWS, the closest analogue is the Generative AI Developer – Professional track (Bedrock Agents, Knowledge Bases); the concepts transfer, but the SDKs and service names differ.
AI-103 is an Associate-level exam and is appreciably harder than the Fundamentals tier. Plan on 40–70 hours of study for developers with some Azure AI background, and more if Foundry and agents are new to you. The exam runs about 100 minutes; expect 40–60 questions in mixed Microsoft formats — multiple choice, multiple response, drag-and-drop ordering, and possibly short case studies — many framed as "choose the best service/approach for this scenario."
The hardest part is breadth: the exam expects working knowledge across model selection, RAG, agents, security, monitoring, vision, speech, and information extraction, all within the Foundry ecosystem. Because the exam and platform are new in 2026, study material is still maturing — lean on the official Microsoft Learn paths, the AI-103 study guide, and hands-on Foundry labs rather than older AI-102 content, which does not cover the agent and Content Understanding material.
New associate exam — Developing AI Apps and Agents on Azure — replacing AI-102. Agent-centric and built around Microsoft Foundry, with five domains spanning planning/management, generative & agentic solutions, computer vision, text analysis, and information extraction. English version updated April 16, 2026; launched as beta and moving to general availability in 2026.
The Azure AI Engineer Associate exam that AI-103 replaces. Structured around implementing Azure AI services (Vision, Language, Speech, Document Intelligence, Azure OpenAI) and generative-AI solutions, without the agent-first, multi-agent, and Content Understanding focus. Retires June 30, 2026.
AI-103 (Microsoft Azure AI Apps and Agents Developer Associate) is a a moderately difficult exam expecting practical hands-on experience plus solid understanding of best practices Associate-level exam. Most candidates need 80–150 hours of study spread over 6–12 weeks for associate-level exams. Most candidates who score consistently above the passing threshold on practice exams pass on their first attempt.
Most candidates need 80–150 hours of study spread over 6–12 weeks for associate-level exams. Time-to-pass varies widely by prior experience. Engineers with hands-on production experience in the underlying technology typically need less; candidates new to the platform should plan toward the upper end of that range.
AI-103 is a recognized credential in the Azure ecosystem and signals validated knowledge to employers, recruiters, and clients. Whether it is worth the time and fee for you depends on your role and goals — it tends to pay off most for cloud engineers, architects, and consultants who work with Azure day-to-day or want to move into roles that do.
The passing score for AI-103 is 700 / 1000. The exam contains 50 questions and lasts 1 hr 40 min.
The AI-103 exam fee is $165 USD. Fees are set by Azure and may vary by region; always confirm the current price on the official Azure certification page before booking.
Microsoft role-based certifications expire after 1 year but can be renewed for free via an unproctored online assessment on Microsoft Learn, starting 6 months before expiration.
Yes. You can take the exam online (proctored via the provider's secure browser, available 24/7 in most regions) or at an in-person Pearson VUE test center during business hours. Both formats use the same questions, time limit, and passing score.
CertLabPro provides 15 study modes across the practice question bank for AI-103. The exam-simulation mode mirrors the real exam: 50 questions in 1 hr 40 min, with the same passing threshold of 700 / 1000. Browse mode lets you read every Q&A statically.