Agents are all the buzz in the AI world – but do we really agree on what an “AI agent” is? In this session, Brian will demystify the concepts of agents and agentic behavior in modern AI development. You’ll get clear definitions and examples so that “agentic” is more than just a buzzword and how things like MCP servers fit in to the puzzle. Brian explores how agentic AI differs from traditional automation, and why it matters for developers building intelligent applications.

We’ll also dive into orchestration frameworks that bring agents to life. From open-source libraries to cloud services, you’ll see how tools like Semantic Kernel enable AI systems to plan, act, and work autonomously toward goals. Through practical demos, Brian will show you how to build and integrate your own AI agents into applications – responsibly and effectively. Whether you’re curious about autonomous GPT-style assistants or just want to clarify the terminology, this talk will leave you confident in using AI agents in your projects.

You will learn:

• What “agentic behavior” really means in an AI context, and how it differs from simply calling something an “agent.”
• How orchestration frameworks (e.g., Semantic Kernel, LangChain) enable multi-step reasoning and tool use for AI agents.
• Practical patterns to build and integrate AI agents into your apps today – and how to avoid the hype and pitfalls as this technology evolves.