Unfortunately, for the past couple of years, Java developers have felt like second-class citizens in the AI space.
While Python has dominated the headlines, we’ve been left wondering if we’ll need to rewrite our entire stack to create even a mediocre autonomous agent.

However, there is good news: the ‘experimental phase' is coming to an end and soon we will enter into ‘production phase’.
This is when the JVM typically takes over and eats everyone’s lunch.

In this session, we’ll be skipping over the fluff and diving into how Java is finally making its move.
We are going to take a top-level look at three libraries that are changing the rules of the game — each offering a different ‘vibe’:
- Spring AI: for when you want enterprise-grade sleep-at-night stability while leveraging the Spring ecosystem you already know.
- Koog (JetBrains): a Kotlin-native powerhouse that helps agent logic feel clean, reactive, and — dare I say it — fun to write.
- Embabel: a clever newcomer that is helping to simplify prompt engineering by focusing on ‘intent’ rather than simply producing a bunch of messy prompts.

We will give a brief "vibe check" to demonstrate how each handles some of the more complex aspects of memory and calling tools without getting lost in 500-page slides.
If you are tired of "python envy" and want to start creating actual AI agents using a stack that doesn't fail under load, this presentation is for you!