KulenDayz2026

28/08/2026, IT Park
Osijek, Croatia

Microsoft AI Workshop [V8]

This workshop will introduce attendees to many of AI and Machine Learning services and tools in the Microsoft world, while providing for hands-on experience. Contents: 1. AI and Machine Learning basics; 2. Introduction to Generative AI, Small and Large Language Models, tokenisation, vectorisation, Agents, Model Context Protocol (MCP), Agent2Agent Protocol (A2A), Retrieval Augmented Generation (RAG), Fine Tuning and more; 2. Introduction to Agents, Model Context Protocol (MCP), Agent2Agent Protocol (A2A), Retrieval Augmented Generation (RAG), Fine Tuning and more; 3. Introduction to Agents using Copilot Studio, Agents and Assistants using Foundry, integration with Microsoft 365 Copilot, etc.; 4. Introduction to Microsoft Foundry (formerly Azure AI Foundry / Azure AI Studio), covering LLMs like GPT models and other Model-As-A-Service options, as well as Agents, the Foundry Agent Service, development tools, extensions and Foundry Local; 5. Introduction to other Microsoft Foundry (AI) Services, covering vision, speech, language and other services; 6. Introduction to Machine Learning and ETL technologies: Microsoft Fabric, Azure Databricks, Azure Machine Learning, etc.; 7. Architecting AI technologies in real life; 8. Q&A. This is a beginner level workshop into AI/ML (no pre-requisite of AI/ML knowledge, but knowledge of programming is expected). Machine learning is a data science technique that allows computers to use existing data to forecast future behaviour, outcomes, and trends. By using machine learning, computers learn without being explicitly programmed. Forecasts or predictions from machine learning can make apps and devices smarter. For example, when you shop online, machine learning helps recommend other products you might want based on what you've bought. When your credit card is swiped, machine learning compares the transaction to a database of transactions and helps detect fraud. And when your robot vacuum cleaner vacuums a room, machine learning helps it decide whether the job is done. Microsoft Fabric, Azure Databricks (and generically Spark, using Python, R, Scala or SparkSQL), and Azure Machine Learning, are cloud services that you can use to train, deploy, automate, and manage machine learning models, as well as process ETL operations on your data, all at the broad scale that the cloud provides. Foundry (AI) Services (formerly AI Services / Cognitive Services) are APIs/SDKs/services available to help developers build intelligent applications without the need for AI or data science skills/knowledge. Azure AI Services enable developers to easily add AI features such as emotion and video detection; facial, speech, and vision recognition; and speech and language understanding – into their applications. The goal of Foundry Services is to help developers create applications that can see, hear, speak, understand, and even begin to reason. Microsoft Foundry Agents, Microsoft 365 Copilot Studio (formerly Power Virtual Agents), and others, use these services integrated to provide a conversational experience that tries to resemble human interaction. Bots can be deployed in channels such as Teams, Slack, Skype, Facebook, Alexa, IoT, or any other you can imagine. Agents can use multiple frameworks for the orchestration layer, such as Microsoft Agent Framework (formerly Semantic Kernel and Autogen), Langchain and even the Bot Framework. Attendees that want to try examples during the workshop need to have: • Laptop; • Have an Azure subscription. You can either set up a free trial at https://azure.com/free, use an Azure Pass free trial, use a paid subscription, or if you have a student account (e.g. at a university), start a student subscription at https://azure.microsoft.com/free/students ; • For Microsoft Fabric, have an active Microsoft 365 subscription with the possibility of starting a free Fabric Trial (this may be limited by your organisation's policies); • Optional: Your preferred development IDE (e.g. the free Visual Studio 2025 Community Edition or Visual Studio Code) already installed, to access online webservices; Note: For AI services and APIs, C# and Python examples will be covered, but equivalent examples exist in other programming languages.