Building LLM Agent Systems with AutoGen Training Course

Foundations of LLM Agent Systems

• Core concepts of LLM agents and multi-agent architectures
• Overview of the AutoGen framework and its ecosystem
• Exploration of agent roles: user proxy, assistant, function caller, and others

Setup and Configuration of AutoGen

• Preparing the Python environment and installing dependencies
• Basics of AutoGen configuration files
• Integration with LLM providers (OpenAI, Azure, local models)

Architecting Agents and Assigning Roles

• Analyzing agent types and conversational patterns
• Establishing agent objectives, prompts, and directives
• Implementing role-based task delegation and control flows

Integrating Function Calls and Tools

• Registering functions for agent utilization
• Executing functions autonomously and in collaboration
• Linking external APIs and Python scripts to agents

Managing Conversations and Memory

• Tracking sessions and maintaining persistent memory
• Handling inter-agent messaging and token processing
• Controlling conversation context and historical data

Developing End-to-End Agent Workflows

• Constructing multi-step collaborative tasks (e.g., document analysis, code review)
• Simulating user-agent dialogues and decision-making chains
• Debugging and optimizing agent performance

Application Scenarios and Deployment

• Internal automation agents: research, reporting, and scripting
• External-facing bots: chat assistants and voice integrations
• Packaging and releasing agent systems for production environments

Wrap-up and Future Directions