AI applications are getting smarter. Not just because of better models, but because of better collaboration. The future is multi-agent, where multiple intelligent agents work together, share context, and coordinate to achieve complex goals.

In this talk, we’ll explore how to build a multi-agent system using Koog, from the first lines of code to a production-ready setup. You’ll learn how to design agents with clear roles, manage their memory and context, and enable communication between sub-agents so they can collaborate effectively.

We’ll also look at how Koog helps you organize your code, orchestrate conversations between agents, and make your system scalable and maintainable.

By the end of this session, you’ll understand how to move from a single-agent demo to a robust, multi-agent architecture powered by Koog.