Belgrade MUG #7: Building AI Agents with MongoDB (Skill Badge Workshop)

2026 has been widely called the year of AI agents, and this workshop will help you understand why.

Join us for a hands-on technical workshop where we will explore how modern AI agents work and how to build them from scratch using today’s most powerful tools.

During the session, you will learn the core building blocks of AI agents such as reasoning, tools, and memory. We will also explore different agent architectures and practical use cases for agentic systems.

The workshop culminates in a hands-on implementation where you will build a technical documentation agent that uses:
• MongoDB as the agent’s memory and knowledge store
• Claude 3.5 Sonnet / Gemini 2.0 / GPT-4o as the agent’s reasoning engine
• LangGraph to orchestrate the agent’s workflow

No local setup is required — everything will run in a cloud environment and can be accessed directly from your browser.

You will be provided with a GitHub repository containing Jupyter Notebook templates, and during the workshop you will replace pseudocode with working implementations.

By the end of the workshop, you will have built your own AI agent capable of answering documentation questions using MongoDB-powered memory.

⏱ Agenda

|18:00 PM - 18:15 PM| - Gathering, networking, food
|18:15 PM - 20:15 PM| - AI Agents Workshop
|20:15 PM - 20:30 PM| - Skill Badge Test
|20:30 PM - 21:00 PM| - Networking, food, swags...

🧠 What you will learn
During this workshop you will:
• Understand the fundamental concepts behind AI agents
• Learn when and how to use agentic architectures
• Build a MongoDB-powered AI agent using LangGraph
• Add memory and retrieval capabilities to your agent
• Learn how to design production-grade agentic workflows