Rod Johnson, the creator of Spring Framework, has launched Embabel. A framework for building AI agents on the JVM. He's called it his most important project since Spring, and after diving in, I can see why.

In this video, we take a first look at Embabel, a framework for authoring agentic flows that seamlessly mixes LLM interactions with code and domain models. We explore the core concepts like actions, goals, and plans (GOAP) and then build a blog writing agent from scratch using Spring Boot, learning about type-safe domain models, personas, multi-model support, and the shell agent starter along the way.

- Understand Embabel's core concepts: Actions, Goals, Plans, and Goal-Oriented Action Planning (GOAP)
- Set up a new Embabel project from scratch with Spring Boot (no templates needed)
- Build a blog writing agent with draft and review actions using type-safe Java records
- Configure multiple LLM models (e.g., GPT-5 Mini for drafting, GPT-5 for reviewing)
- Use Embabel's persona system for reusable role definitions across actions

#java #embabel #springai #aiagents #springboot #rodjohnson #jvm #programming

📦 GitHub Repository (Blog Agent): https://github.com/danvega/blog-agent
🔗 Embabel GitHub: https://github.com/embabel
📖 Embabel Documentation: https://docs.embabel.com/embabel-agent/guide/0.4.0-SNAPSHOT/

👋🏻Connect with me:
Website: https://www.danvega.dev
Twitter: https://twitter.com/therealdanvega
Github: https://github.com/danvega
LinkedIn: https://www.linkedin.com/in/danvega
Newsletter: https://www.danvega.dev/newsletter

SUBSCRIBE TO MY CHANNEL: http://bit.ly/2re4GH0 ❤️

Dan Vega

A hands-on getting started tutorial for Embabel, a JVM framework for building AI agents created by Rod Johnson (founder of Spring Framework). Covers core concepts including goal-oriented action planning (GOAP), actions, goals, and plans. Walks through building a blog-writing agent from scratch using Spring Boot 3.5, demonstrating type-safe domain models with Java records, multi-model configuration (GPT-4o-mini for drafting, GPT-4o for reviewing), personas as reusable prompt contributors, and writing output to markdown files. Also shows the shell interface for inspecting agents, listing models, and debugging LLM interactions with verbose output.

Building AI Agents in Java with Embabel (Getting Started)