Description 

LLMs (Large Language Models) are no longer a novelty, they can chat, analyze, summarize, and generate content. But how do you build a real system that works for you, not just answers your questions? 

In this hands-on workshop, you’ll learn how to build autonomous AI Agents, and more importantly, how to think like one: identify a problem, break it down into distinct roles, and orchestrate a team of LLMs to execute a complex, intelligent process. 

We’ll explore the difference between “an LLM that replies” vs. “an Agent that works,” and guide you through the process of building your own multi-agent system that actually gets work done. 

By the end of the day, you’ll have a new mindset: not just how to ask questions, but how to recognize needs, design intelligent workflows, and outsource operations to AI. 

What You’ll Gain: 

• A solid foundation in building real AI Agent workflows 

• Experience with state-of-the-art tools like LangChain and Autogen 

• Strategies for orchestrating multi-agent systems 

• Knowledge of best practices in reasoning, grounding, and memory in LLMs 

• Your own custom-built prototype of an AI Agent or Agent team 

• Confidence to continue developing AI-powered systems for real-world tasks 

Main Topics 

• Foundations & Concepts 

• Solving real-world problems with multi-agent teams 

• Core concepts: Introduction to ML, where psychology meets machines, neural networks 

• What are LLMs? Different types and models in the market 

• Understanding key limitations of current LLMs 

• Building a Smart Agent: Advanced LLM Techniques 

• Chain of Thought: When step-by-step reasoning changes everything 

• Advanced reasoning techniques 

• RAG (Retrieval-Augmented Generation) for grounding models in reliable data 

• Practical examples and guided exercises 

• Exercise: Training the model to “think” before answering 

• Introduction to AI Agents 

• What is an Agent, and why it’s not “just a chatbot” 

• Defining tools, roles, and task boundaries 

• Agent design patterns: autonomy, delegation, monitoring 

• Mapping a real-world challenge into an Agent-based solution 

• Exercise: Designing an Agent for a specific use case 

• Hands-On: Build Your First AI Agent 

• Using frameworks like LangChain, Autogen, and memory tools 

• Creating a simple autonomous Agent to solve a defined scenario 

• Connecting tools, crafting instructions, managing memory 

• No-Code Solutions 

• Overview of no-code platforms for Agent development 

• Trade-offs: flexibility vs. ease-of-use 

• When to use no-code, and when to go full-stack 

• Multi-Agent Collaboration: Who Does What? 

• Designing coordinated teams of AI Agents 

• Roles, responsibilities, control and oversight 

• Systemic thinking in AI workflows 

• Exercise: Defining and running a full Agent team 

• Final presentations of participants’ creations 

Target Audience:  

• Software developers 

• AI/ML practitioners 

• Product & automation engineers 
• Tech leads looking to implement intelligent, scalable processes