Microsoft Azure AI Apps and Agents Developer Associate (AI-103)

Choose Foundry services, set up AI solutions, and manage, monitor, secure and govern them responsibly.

• Choosing LLMs, small language models, multimodal models and Foundry Tools per task
• Choosing Foundry services for generation, grounding, vector search, agent workflows and multimodal processing
• Choosing retrieval and indexing methods, and memory/tool/knowledge integration for agents
• Designing Azure infrastructure and deployment options for AI apps and agents
• Configuring model and agent deployments; integrating Foundry projects with CI/CD
• Managing quotas, scaling, rate limits and cost footprints
• Monitoring model performance, drift, safety events and grounding quality
• Monitoring ingestion quality, search index health and relevance
• Managed identity, private networking, keyless credentials and role policies
• Safety filters, guardrails, risk detection and content moderation
• Responsible AI instrumentation: evaluators, safety evaluations, explanation tooling
• Auditing via trace logging, provenance metadata and approval workflows
• Governing agent behaviour with oversight modes, constraints and tool-access controls

Free Plan and manage an Azure AI solution practice questions →Plan and manage an Azure AI solution study guide →

Build generative applications and agents on Foundry, then optimise and operationalise them.

• Deploying and consuming LLMs, small models, code models and multimodal models
• Implementing retrieval-augmented generation in an application
• Designing workflows, tool-augmented flows and multistep reasoning pipelines
• Evaluating models and apps for fabrication, relevance, quality and safety
• Integrating generative workflows via Foundry SDKs and connectors
• Defining agent roles, goals, conversation tracking and tool schemas
• Agents combining retrieval, function-calling and conversation memory
• Integrating agent tools: APIs, knowledge stores, search, Content Understanding, custom functions
• Orchestrated multi-agent solutions
• Autonomous and semi-autonomous workflows with safeguards and approval flows
• Prompt engineering and model parameter tuning
• Model reflection, chain-of-thought evaluation and self-critique loops
• Observability: tracing, token analytics, safety signals, latency breakdowns
• Orchestrating multiple models, flows, or hybrid LLM and rules engines

Free Implement generative AI and agentic solutions practice questions →Implement generative AI and agentic solutions study guide →

Generate and edit images and video, build multimodal understanding workflows, and apply responsible AI to visual content.

• Generating images from text prompts and reference media
• Generating video from text prompts and reference media
• Image-editing workflows: inpainting, mask-based edits, prompt-driven modification
• Editing generated video
• Analysing visual context with multimodal models
• Concise and detailed captioning for single and multiple images
• Question-answering grounded in visual evidence
• Alt-text and extended image descriptions aligned to accessibility guidelines
• Azure Content Understanding in Foundry Tools for visual characteristics
• Video analysis workflows and segment interpretation
• Single-task and pro-mode Content Understanding pipelines
• Identifying objects, components and regions in images and video
• Filtering unsafe or disallowed visual content
• Detecting indirect prompt injection via text embedded in images
• Visual policy rules: watermarks, prohibited symbols, brand usage

Free Implement computer vision solutions practice questions →Implement computer vision solutions study guide →

Apply language models to text analysis and build speech solutions for agentic interactions.

• Extracting entities, topics, summaries and structured JSON via generative prompting and Foundry Tools
• Detecting sentiment, tone, safety issues and sensitive content
• Translation with Azure Translator in Foundry Tools and LLM-powered flows
• Customising outputs for domain tasks such as compliance summarisation
• Speech-to-text and text-to-speech for agentic interactions
• Speech as an agent modality, including custom speech models
• Multimodal reasoning from audio inputs
• Speech translation using language models and Foundry Tools

Free Implement text analysis solutions practice questions →Implement text analysis solutions study guide →

Build retrieval and grounding pipelines and extract structured content from documents.

• Ingesting and indexing documents, images, audio and video
• Semantic, hybrid and vector search for grounding
• Enrichment with custom and built-in skills for text, images and layout
• RAG ingestion flow including OCR
• Connecting retrieval pipelines to workflows and agent tools
• Multimodal extraction combining OCR, layout analysis and field extraction
• Clean grounded representations for agents and RAG via Content Understanding
• Analyzers producing structured or markdown output for downstream reasoning

Free Implement information extraction solutions practice questions →Implement information extraction solutions study guide →