Microsoft AI-103 Training Course – Developing AI Apps and Agents on Azure
- Price
- Duration
- Number of hours
Our sessions are guaranteed from 1 registered participant (except in cases of force majeure).
Description of the Microsoft AI-103 Training Course
This Microsoft AI-103 – Developing AI Apps and Agents on Azure course prepares developers and engineers to design, secure and deploy AI applications and agents on Azure with Microsoft Foundry, the unified platform for model management and agent orchestration. You will learn to build generative and agentic AI solutions (Azure OpenAI, RAG with Azure AI Search), handle vision, language and information extraction, and apply production and responsible-AI best practices. Through hands-on labs, you will develop real AI apps and agents and master their secure implementation on Azure.
Format
Remote (recorded sessions).
It is possible to customise the training content to meet the needs of your professional project.
GOOD TO KNOW
This training course includes many exercises (60% practice) for better learning. Sessions are guaranteed from 1 registered participant (except in cases of force majeure). A preliminary interview takes place between the participant and/or a company representative to properly account for the participant’s profile (level, needs, professional context, challenges…).
Assessment: during the training, the trainer evaluates the pedagogical progress of participants through MCQs, role-play scenarios and practical exercises. Participants receive a certificate of skills validation at the end of the training.
This training course is part of our Artificial Intelligence Training Courses. Explore our other AI training courses to fully leverage machine learning, LLMs and generative AI.
Objectives of the Microsoft AI-103 Training Course
Upon completion of the training, the participant will be able to:
- Plan, provision and manage an Azure AI solution in Microsoft Foundry: service selection, deployment options, networking and private endpoints, RBAC, quotas, cost management, monitoring (Application Insights) and responsible AI.
- Design and implement generative AI and agentic solutions: Foundry projects, prompt engineering, Azure OpenAI models, RAG pipelines with Azure AI Search, multi-agent orchestration, conversation memory and response evaluation.
- Develop computer vision solutions: image analysis, OCR, image and video generation from prompts, multimodal processing.
- Implement LLM-first text analysis solutions: entity extraction, classification, summarisation, translation and speech recognition with Azure AI Language and Speech.
- Build information extraction pipelines with Azure AI Document Intelligence and Content Understanding (documents, images, audio, video) to power RAG scenarios.
Prerequisites
- Application development experience in Python.
- Familiarity with general AI, generative AI concepts and Azure services.
- Proficiency with REST APIs and SDKs to integrate Azure AI services.
- Knowledge of Azure fundamentals (portal, resources); AI-900 or equivalent recommended.
- Because each participant is unique, a personalised interview with our expert allows us to design a training course perfectly aligned with their objectives, level and professional challenges.
Target Audience
- Developers and Azure AI engineers building apps and agents with Microsoft Foundry.
- Python developers integrating AI into business applications.
- Technical teams aiming to master generative AI, agents, RAG and Azure AI services.
Detailed Programme of the Microsoft AI-103 Training Course
Plan and manage an Azure AI solution
Azure AI service selection and setup in Microsoft Foundry, networking, private endpoints and security, RBAC (Cognitive Services OpenAI User vs Contributor), quotas, cost management, monitoring with Application Insights, responsible AI principles and CI/CD integration.
Implement generative AI and agentic solutions
Identify business use cases suited to pre-built models. Analyse documents using the Read, Layout or General Document models. Process specialised documents (financial, tax, identity documents).
Implement computer vision solutions
Image analysis, OCR, image and video generation from prompts and reference media, editing workflows (inpainting, masks), spatial video analysis.
Implement text analysis solutions
Entity extraction, classification and summarisation with Azure AI Language, speech recognition and processing with Azure AI Speech, translation, LLM-first approach.
Implement information extraction solutions
Azure AI Document Intelligence (prebuilt and custom models), Content Understanding across documents, images, audio and video, ingestion and structuring to feed RAG pipelines.
Capstone project
Building an end-to-end AI solution on Microsoft Foundry combining agent(s), multimodal RAG, governance and security.
FAQ – Microsoft AI-103 Azure AI Engineer Training
What is the Microsoft AI-103 certification?
AI-103 (“Developing AI Apps and Agents on Azure”, the Azure AI Apps and Agents Developer Associate certification) validates the ability to build, deploy and manage AI apps and agents on Azure with Microsoft Foundry: generative and agentic AI, Azure OpenAI, RAG with Azure AI Search, vision, text analysis and information extraction (Document Intelligence, Content Understanding).
What is the difference between AI-103 and AI-102?
AI-103 replaces AI-102 (Azure AI Engineer Associate), which is scheduled to retire on 30 June 2026. AI-103 refocuses the exam on Microsoft Foundry and makes generative AI and agentic solutions (agents, multi-agent orchestration, RAG, conversation memory) the main domain, whereas AI-102 was organised service by service. The MFE-IT course follows the new AI-103 syllabus.
What are the prerequisites for the AI-103 course?
Python development experience, knowledge of REST APIs and SDKs, and familiarity with general AI, generative AI and Azure services. Knowledge of AI-900 or equivalent is recommended. The MFE-IT course includes targeted refreshers at the start of the session to level up participants.
How long does it take to prepare for the AI-103 certification?
Our Microsoft AI-103 course runs over 4 days (24 hours) in a fully tailored format. This covers the five exam domains, with hands-on labs on Microsoft Foundry, agents, generative AI and RAG.
Is this Microsoft AI-103 course eligible for CPF or OPCO funding?
No. MFE-IT does not directly manage CPF or OPCO files and our organisation is not Qualiopi-certified. This course is therefore intended only for companies that fund it directly. In return, you benefit from a 100% tailored format: preliminary interview, content adapted to your business context, sessions guaranteed from 1 registrant (except in cases of force majeure), a maximum of 3 participants per session and 30 days of post-training email support.
Would you like to know about upcoming sessions?
Would you like to schedule this Microsoft AI-103 training course on a specific date? Contact us by email or by filling out the contact form.