AI-103: Azure AI App and Agent Developer - Complete Course

Name: AI-103: Azure AI App and Agent Developer - Complete Course
Rating: 4.5 (239 reviews)

Master the 2026 Agentic Shift: Build, deploy, and scale production-ready AI agents using Azure, Python, and RAG.

Bestseller

Highest Rated

Created byLuke Ginn

Last updated 7/2026

English

What you'll learn

Master Agentic AI on Azure: Dive into cutting-edge topics like multi-agent orchestration, reasoning loops, and grounding.
Reinforce Azure Knowledge: Strengthen understanding of AI-103 topics (agent orchestration, grounding, memory, Entra Agent ID, and more)
Design Scalable Multi-Agent Workflows: Learn to architect complex AI systems using the latest frameworks to solve multi-step business problems
Optimize Retrieval-Augmented Generation (RAG): Gain hands-on experience in grounding agents with Azure AI Search and Azure Cosmos DB

Course content

4 sections • 61 lectures • 25h 21m total length

Unit 1: What Is an AI Agent?27:28
Preview: Unit 1 Code Walkthrough: Deploying your first Agent23:05
Essential Guide to Course Resources & Exercises1:29
Unit 2: Microsoft Foundry – The Agent Development Platform25:28
Lab 1: Microsoft Foundry Intro26:34
Unit 3: Entra Agent ID – Agent Identity Deep Dive22:33
Lab 2: Microsoft Foundry Agent Intro20:44
Deploy a Foundry resource, build and deploy a GPT-5 Mini based agent, optimize system instructions, enable tracing and web searching, then publish and grant access via Azure Bot services.
Unit 4: Responsible AI & Safety for Agents20:31
Lab 3: Responsible AI & Safety for Agents21:49
Unit 5: Model Selection – LLMs vs. SLMs for Agent Tasks20:43
Compare large language models and small language models to select the right brain for your agent, balancing cost, latency, and task with routing and hybrid strategies.
Lab 4: LLMs vs SLMs for Agent Tasks18:32
Unit 6: Agentic RAG – Dynamic Retrieval for Agents22:11
Explore agentic rag for grounding AI with your data, using vector embeddings and searches via Azure AI Search and Bing, contrasting static and agentic approaches, with Fabric and Cosmos DB.
Lab 5: Agentic RAG19:43
Unit 7: Advanced Grounding – Fabric & Cosmos DB21:53
Unit 8: Context Management – System Instructions18:45
Design robust system instructions that define the agent's persona, boundaries, safety rules, and grounding data, ensuring persistent behavior, tool use, and safe, accurate responses across conversations.
Unit 9: Multi-Agent Orchestration – Microsoft Agent Framework21:05
Lab 6: Multi-Agent Orchestration - Routing23:21
Unit 10: Tool Calling – Connecting Agents to APIs20:10
Lab 7: Tool Calling & Basic MCP25:31
Unit 11: Model Context Protocol (MCP) – Standardized Tool Connectivity27:01
Lab 8: More About MCP Servers23:05
Unit 12: Sequential vs. Concurrent Agent Execution18:57
Unit 13: Human-in-the-Loop (HITL) – Approval Gates21:56
Master human-in-the-loop design with approval gates and escalation, pausing agent actions for human judgment. Integrate logic apps, manage timeouts, and monitor decisions with Foundry Trace for secure, high-stakes workflows.
Unit 14: Foundry Trace – OpenTelemetry Debugging23:29
Lab 9: Foundry Trace11:11
Unit 15: Performance Evaluation – Metrics That Matter24:55
Unit 16: Agent Memory – Short-Term and Long-Term State20:09
Lab 10: Agent Memory12:32
Unit 17: Multimodal Integration – Vision, Speech & Document Intelligence24:30
Unit 18: Input Transformation – NLP Preprocessing for Agents17:05
Unit 19: System Instructions – Programmatic Prompt Construction13:12
Unit 20: Complete Agent Solution – Integration & Exam Readiness26:00

Unit 1: Foundation – Your First Agent Call23:05
Set up a virtual environment, log in with Azure CLI, deploy AI Foundry resources with Bicep, and call a GPT 4.1 mini model via the OpenAI SDK.
Unit 1: Foundation - Your First Agent Call (Repeat due to Udemy bug)23:05
Read Before You Start: Deployment & Model Notes1:20
Special Unit: Bicep Deployment, Tips, Tricks & Advice49:20
Unit 2: Content Safety - Input and Output Filtering35:06
Unit 3: System Instructions – Dynamic Prompt Construction38:48
Unit 4: Model Selection – LLM vs SLM Routing38:29
Unit 5: LLM Model Parameters (Temperature, etc.)36:18
Explore how to tune large and small language model parameters, including max past messages, max tokens, temperature, and top P, via a config.yaml; assess impact on cost, latency, and routing.
Unit 6: Azure AI Search (Part 1) - Basic Search and Retrieval23:29
Unit 7: Azure AI Search (Part 2) - Vector Search and Semantic Ranking27:58
Unit 8: Azure AI Search (Part 3) - Agentic RAG23:10
Deploy Azure AI Search resources with Foundry, content safety, and llms (large, small, embedding) in sequence, then enable grounding results for agentic rag to handle refund queries.
Unit 9: Tool Calling – REST API Integration43:41
Explore tool calling and REST API integration in unit 9 of the Azure AI App and Agent Developer course, enabling seamless access to external services from AI agents.
Unit 10: Model Context Protocol (MCP) – Standardized Tool Connectivity38:30
Unit 11: Multi-Agent - Handoff Pattern26:22
Unit 12: Multi-Agent – Magentic (Manager) Pattern24:46
Learn the magentic manager pattern in this unit: a hierarchical multi-agent approach where a manager routes requests to specialized sub-agents (refund, products, accounts) with no state transfer.
Unit 13: Sequential vs Concurrent Agent Execution29:13
Unit 14: OpenTelemetry Debugging39:28
Unit 15: Agent Memory – Short-Term Session Memory (Redis)37:01
Unit 16: Agent Memory – Long-Term User Memory (Cosmos DB)32:27
Unit 17: Multimodal Integration – Vision, Speech, Document Intel28:22
Unit 18: Input Transformation – NLP Preprocessing26:22
Unit 19 : Secure Secret Management with Azure Key Vault25:52
Unit 20: Complete Agent Integration – Production-Ready System41:17

Theory: Unit 1: The Absolute Basics – What Is an AI Integration Engineer?20:30
Learn to build productionized AI systems on azure by integrating containers, vectors, and events. Use these pillars to enable scalable inference, semantic search, and decoupled workflows.
Theory: Unit 2: Storing Containers – Azure Container Registry (ACR)20:41
Theory: Unit 3: Serverless Containers – Azure Container Apps18:15
Learn how Azure Container Apps run container images from Azure Container Registry as serverless apps with autoscaling, revisions, environment variables, secrets, health probes, and KEDA-based scaling for AI services.
Code Walkthrough: Unit 1: Azure Container Apps with Hello World FastAPI21:52
Build a basic FastAPI hello world API with get and post endpoints validated by Pydantic, including health checks, and deploy it to Azure Container Apps via a Docker image.
Code Walkthrough: Unit 2: Azure Container Apps with Application Factory FastAPI29:26
Build a configurable FastAPI app with the application factory pattern using createApp, include pydantic validation and endpoints, then deploy and test on Azure container registry and container apps.
Code Walkthrough: Unit 3: Azure Container Apps for AI Agents31:50
Build an Azure Container Apps api that calls OpenAI chat completions via FastAPI, using the application factory pattern, an async client, environment-driven settings, and readiness probes.

Requirements

Basic Azure familiarity is helpful but not required; the course teaches AI-103 concepts from the ground up with clear explanations.
A willingness to learn and engage with technical material to build confidence for the actual AI-103 exam.

Description

AI-103: Azure AI App and Agent Developer - Complete Course

Key Benefits

End-to-End Implementation: Go beyond theory with comprehensive lectures and hands-on code walkthroughs covering Multi-Agent Orchestration, Agentic RAG, Foundry Trace, and Entra Agent ID.
Architectural Mastery: Understand the 'why' behind agent design patterns with deep dives into the Microsoft Agent Framework from official documentation.
Build Real-World Solutions: Move from concept to code by building functional agents that utilize Grounding, Memory, and advanced Tool Calling.
Pass the AI-103 Confidently: Complement your exam practice with the technical depth and practical experience required to master Microsoft’s latest AI-103 certification.

Are you ready to lead the 2026 shift from model-centric APIs to entity-centric orchestration?

The AI-103 exam is challenging because it requires more than just knowing concepts—it requires the ability to architect and implement complex agentic workflows. This course is designed to take you from the fundamentals of Azure AI Foundry to the cutting edge of multi-agent reasoning and observability.

Updated for the latest 2026 syllabus, this lecture-based course provides the technical "missing link" between documentation and deployment. While our sister course provides the practice questions, this course provides the hands-on mastery needed to build the solutions those questions describe.

What You Will Build and Master:

Through detailed modules and technical walkthroughs, you will cover every objective domain tested on the AI-103 exam:

Foundry Management & Governance: Set up secure environments using Entra Agent ID and implement Red Teaming for agent safety.
Generative AI & Advanced Grounding: Implement Agentic RAG and integrate with Microsoft Fabric for high-context data retrieval.
Multi-Agent Orchestration & Reasoning: Program complex handoffs and manager patterns using the Microsoft Agent Framework and MCP.
Agentic Observability & Memory: Configure Foundry Trace for debugging and manage long-term state persistence.
Agentic Enhancement: Extend your agents with Vision/Speech APIs and connect to enterprise data via Logic Apps.

How This Course Will Get You Certified:

Deep Technical Lectures: We break down complex orchestration patterns into digestible, visual lessons so you understand the logic behind the "Agentic Shift."
Code-First Approach: Every major concept is accompanied by a code walkthrough, ensuring you can implement Entra Agent ID or Foundry Trace in your own projects.
Bridge the Gap: This course is the perfect companion to the "AI-103 Practice Exams" course. Learn the how here, then test your speed there.
Future-Proof Your Career: Focus on the latest 2026 standards for AI development, moving away from simple prompting into the world of autonomous, tool-calling agents.

This Course is Perfect For:

AI Developers seeking AI-103 certification who want to master multi-agent orchestration and observability through technical deep-dives.
AI Architects building enterprise solutions on Microsoft Foundry who need to understand the underlying infrastructure and security of agentic systems.

Requirements:

Basic Azure familiarity is helpful; while we cover concepts from the ground up, a general understanding of cloud services will accelerate your progress.
A developer mindset: You should be ready to engage with code, logic flows, and the technical architecture of the Microsoft Agent Framework.

Who this course is for:

Aspiring AI Agent Developers: Those aiming for AI-103 certification, who want to master multi-agent orchestration, grounding, and observability through lectures and walkthroughs
AI Developers Seeking Certification: Developers with some AI or cloud experience, seeking AI-103 certification to validate their agentic development skills and committed to mastering multi-agent orchestration.

AI-103: Azure AI App and Agent Developer - Complete Course

What you'll learn

Explore related topics

Course content

Theory / Labs32 lectures • 11hr 6min

Code Walkthroughs23 lectures • 11hr 53min

Exam Questions0

Bonus: AI-200 Preview6 lectures • 2hr 23min

Requirements

Description

Who this course is for: