AI-103 Microsoft Certified Azure AI Apps and Agents Developer Associate

This course prepares you for the AI-103 certification exam, covering all official skills measured and their approximate weightings:

Domain 1 Plan and manage an Azure AI solution (25-30%)

• Choose appropriate models (LLMs, small and multimodal models) and Foundry services for generative tasks, grounding, vector search, agent workflows and multimodal processing
• Choose retrieval and indexing methods and memory, tool and knowledge integration services for agents
• Design Azure infrastructure, choose deployment options, configure model and agent deployments, and integrate Foundry projects with CI/CD pipelines
• Manage quotas, scaling, rate limits and cost, and monitor model performance, drift, safety events and grounding quality
• Configure security including managed identity, private networking, keyless credentials and role policies
• Implement responsible AI: safety filters, guardrails, content moderation, evaluators, audit trace logging and agent oversight controls

Domain 2 Implement generative AI and agentic solutions (30-35%)

• Deploy and consume LLMs, small, code and multimodal models, and implement retrieval-augmented generation (RAG)
• Design tool-augmented flows and multistep reasoning pipelines, and evaluate models and apps for fabrications, relevance, quality and safety
• Integrate generative workflows using Foundry SDKs and connectors, and connect an application to a Foundry project
• Define agent roles, goals and tool schemas, and build agents integrating retrieval, function-calling and conversation memory
• Implement orchestrated multi-agent solutions and autonomous workflows with safeguards and approval controls
• Optimize with prompt engineering, model parameters, reflection and self-critique loops, and set up observability with tracing, token analytics and safety signals

Domain 3 Implement computer vision solutions (10-15%)

• Generate images and videos from text prompts and reference media, and configure image and video editing workflows
• Build multimodal understanding: analyze visual context, produce captions, and enable question-answering grounded in visual evidence
• Generate accessible alt-text and image descriptions, and extract visual characteristics with Azure Content Understanding in Foundry Tools
• Implement video analysis and identify objects, components or regions within images and video
• Implement responsible AI for multimodal content: filter unsafe visuals, detect indirect prompt injection in images, and enforce visual policy rules

Domain 4 Implement text analysis solutions (10-15%)

• Extract entities, topics, summaries and structured JSON using generative prompting and Foundry Tools
• Configure detection of sentiment, tone, safety issues and sensitive content
• Translate text using Azure Translator in Foundry Tools or LLM-powered translation flows
• Customize language model outputs for domain tasks such as compliance summarization and domain extraction
• Implement speech-to-text and text-to-speech, custom speech models, multimodal reasoning from audio, and speech translation

Domain 5 Implement information extraction solutions (10-15%)

• Ingest and index documents, images, audio and video, and configure semantic, hybrid and vector search for grounding
• Implement enrichment skills and configure RAG ingestion flow including optical character recognition (OCR)
• Connect retrieval pipelines directly to workflows and agent tools
• Extract information using multimodal pipelines combining OCR, layout analysis and field extraction
• Produce clean, grounded representations and structured or markdown outputs with Content Understanding