AI-103 study guide
AI-103 · Work through your exam, one topic at a time.
AI-103 leads to the Microsoft Certified: Azure AI Apps and Agents Developer Associate credential, an intermediate-level certification for AI engineers and developers who build, manage, and deploy agents and AI solutions that take advantage of Microsoft Foundry. Candidates should already be comfortable developing apps in Python and should understand the capabilities of general AI, generative AI, and Azure services. The role spans planning and managing Azure AI solutions plus implementing generative AI and agentic, computer vision, text analysis, and information extraction workloads, and it involves collaborating with business stakeholders, solution architects, data scientists, DevOps engineers, and cloud security engineers. The assessment is proctored, lasts 120 minutes, and may include interactive components alongside other item types. The AI-103 study guide has 5 sections and about 1,630 words.
01Plan and Manage an Azure AI Solution
Every solution starts with the account decision. A multi-service Azure AI Services resource gives you one endpoint and one key across many APIs, which simplifies secret management and billing - but a few capabilities, particular regions, and container deployments still require a single-service resource.
322 words, about 2 min02Implement Generative AI and Agentic Solutions
You deploy a model (for example GPT-family) into an Azure OpenAI resource and consume it by deployment name, not model name. Deployment types trade cost against latency predictability: Global Standard is cheapest with variable latency, Standard is regional, Provisioned reserves throughput for steady high load.
352 words, about 2 min03Implement Computer Vision Solutions
Modern Azure vision work splits into three habits. First, Image Analysis provides structured outputs: caption, dense captions, tags, object detection with bounding boxes, people, smart crops, and OCR via the Read model.
304 words, about 2 min04Implement Text Analysis Solutions
Know which feature answers which question. Sentiment analysis returns document- and sentence-level scores; opinion mining is required to attach sentiment to specific aspects or targets.
323 words, about 2 min05Implement Information Extraction Solutions
Choose the model by document variability, not by document name. Prebuilt models (invoice, receipt, ID document, health insurance card, bank statement, contract) return a defined field schema with confidence scores.
329 words, about 2 min