Artificial Intelligence is reshaping software development, but beneath the buzzwords lie practical mechanics every engineer should understand. Designed specifically for developers, DevOps professionals, and software engineers, this 60-minute deep dive pulls back the curtain on how AI systems actually function under the hood.

In this session, we will break down the core architecture of Large Language Models (LLMs), separating reality from hype. You'll explore how these models process data, whether they truly "learn" or simply predict pattern boundaries, and how modern systems dynamically handle real-time context when hitting training knowledge limits. Finally, we’ll analyse the shift from passive text completion to active execution with "agentic" AI—unpacking how autonomous agents plan, reason, and interface with external toolsets.

In this 60-minute workshop, we'll cover: 

• What is AI, really?
• How do Large Language Models (LLMs) work?
• Do they actually learn?
• What happens when they need information they were not trained on?
• What is “agentic” AI?