Content:

This training is aimed at developers and architects who want to develop modern agent systems for complex automation and AI scenarios based on Autogen and Azure. The training clearly presents the basics and concepts and teaches them using practical examples.

Microsoft AutoGen is a framework for developing agent systems. It enables the implementation of simple chat agents with tools for multi-agent workflows. In combination with deployment on Azure, this results in flexible and scalable AI solutions.

Over the course of the training, you will master the following topics:

– Recap of Artificial Intelligence and Large Language Models:
... - History of artificial intelligence and development
... - Fundamentals of Large Language Models (LLMs)
... - Prompt engineering techniques and best practices
... - Azure OpenAI Service overview

– Introduction to AI Agents:
... - Basic concepts of agents and autonomous systems
... - Agent patterns and workflows
... - Overview of agent frameworks and comparison
... - Use cases and application scenarios

– AutoGen Basics:
... - Framework structure: AgentChat API and components
... - Basic structure of an agent in C#
... - Built-in agents and basic functionalities
... - Integration in ASP.NET Core applications
... - Hands-on: Implementing your first chat agent

– Tool Integration and Function Calling:
... - Tool calling concepts in AutoGen
... - Connecting RESTful APIs and external services
... - Developing and registering custom tools
... - Error handling and retry strategies
... - Hands-on: Agent with external APIs

– Multi-Agent Basics:
... - Multi-agent patterns and orchestration strategies
... - Agent-to-agent communication and protocols
... - Group chat and agent coordination
... - Human-in-the-loop integration and workflows
... - Hands-on: Simple multi-agent system

– Memory and State Management:
... - Session and memory management in AutoGen
... - Persisting agent states
... - Context management and conversation history
... - Performance optimization with large contexts

– Advanced Features (Overview):
... - Swarm and GraphFlow for complex workflows
... - RAG integration for knowledge-based agents
... - Logging and tracing for debugging
... - Custom agent types and extensions

– Security and Testing:
... - Basic security aspects of agent systems
... - Input validation and content filtering
... - Testing approaches for agents and multi-agent systems
... - Unit testing and integration testing

– AutoGen Deployment on Azure:
... - Containerization of AutoGen applications
... - Azure Container Apps deployment and configuration
... - Integration into CI/CD systems and DevOps pipelines
... - Monitoring and logging with Azure Monitor
... - Scaling and performance tuning

– Summary and Best Practices:
... - Single vs. multi-agent design decisions
... - Architecture patterns for production systems
... - Additional resources and community
... - Roadmap and future developments

This course combines theoretical knowledge with practical application and prepares you to successfully develop and operate production-ready agent systems in Azure.

Disclaimer: The actual course content may vary from the above, depending on the trainer, implementation, duration and constellation of participants.

Whether we call it training, course, workshop or seminar, we want to pick up participants at their point and equip them with the necessary practical knowledge so that they can apply the technology directly after the training and deepen it independently.