If you want to understand the current state of the art in generative AI technology, this session is for you.

We will start with LLM Basics: How do Large Language Models (LLMs) work internally (at a high level of abstraction)? How does the API of an LLM look like, and what are its limitations? How do "reasoning models" work?
Then we'll visit the peripherals of a LLM: What is Retrieval Augmented Generation (RAG)? What is "tool calling"?
Before coming to AI Agents answering the questions: What defines an AI Agent? How do current state-of-the-art AI agents function? What's the Model Context Protocol (MCP)?

We will conclude the session with a practical example of how an internal AI agent for a software development team could look like. We will also give away some lessons learned about AI use cases and their pitfalls.