Agentic AI Unleashed: Crafting LLM Applications with AutoGen

. Prerequisites - Python programming

This 1-day workshop, designed for developers, data scientists, and AI enthusiasts, will help you understand and harness the power of agentic AI systems using AutoGen v0.4. 

Through a mix of hands-on exercises and practical demonstrations, you’ll learn how to build, manage, and deploy multi-agent applications powered by Large Language Models (LLMs). 

By the end of the course, you'll gain a solid foundation in AutoGen’s layered architecture, master asynchronous communication between agents, and explore real-world use cases and best practices for developing scalable and intelligent LLM-driven applications.

1. Introduction to LLM Applications and AutoGen v0.4

• Overview of Large Language Models (LLMs): Understanding their capabilities and applications.​ 
• Introduction to AutoGen v0.4: Exploring its features, architecture, and how it simplifies the development of agentic AI systems.  

2. Core Concepts and Components of AutoGen

• Understanding the Layered Framework:
  • Core Layer: Event-driven architecture supporting dynamic workflows.
  • AgentChat API: Building task-driven agents with high-level APIs.
  • Extensions: Integrating custom agents, tools, and memory modules for enhanced functionality.
• Asynchronous Messaging: Implementing event-driven and request-response interaction styles.​ 

3. Building Your First Multi-Agent Application

• Defining Agents: Creating Assistant and User Proxy agents.​ 
• Establishing Agent Communication: Setting up asynchronous messaging between agents. 
• Implementing a Sample Application: Developing a simple multi-agent system to solve a specific task.​ 
• Observability and Debugging Tools: Utilizing built-in metric tracking and message tracing for real-time monitoring.​ 

4. Case Studies and Best Practices

• Real-World Applications: Examining successful implementations of AutoGen in various industries.​
• Best Practices: Guidelines for designing efficient and scalable LLM applications using AutoGen.​
• Challenges and Solutions: Addressing common challenges faced during development and their solutions.​
• Q&A

The workshop is intended for:

• software developers
• data scientists
• data engineers
• people with programming background/inclination who want to learn about AI programming.