In the third episode of IntelliJ IDEA Tech Talks, I talk to Ronald Dehuysser – creator of JobRunr, about the ClawRunr project. Ronald shows us how to use ClawRunr as a personal AI assistant and shares stories of how he uses it himself.
Description
What if you had a personal assistant who could check your emails, schedule tasks, prepare a legal dossier, and send a thank-you note, all from your phone while you’re on a ski slope?
ClawRunr is an open-source Java-based personal assistant that combines AI with background job scheduling. You chat with it in natural language, it figures out what to do, and Spring AI and JobRunr handle the rest.
Ronald Dehuysser, the creator of JobRunr, walks us through setting up ClawRunr, from choosing your LLM provider and connecting MCP servers to scheduling recurring tasks using live demos. He also shares what he learned taking it from a vibe-coded idea to an extensible demo app: why solid architecture matters for agent-style development, and why handing off as much as possible to the AI beats coding everything yourself.
Chapters
- 00:00:00 – Introduction
- 00:00:32 – What is ClawRunr
- 00:02:23 – Setting up ClawRunr
- 00:07:33 – From vibe-coded idea to demo app
- 00:18:34 – Live demo with email triage, court case and scheduling
- 00:29:57 – Recurring tasks and scheduling with JobRunr
- 00:34:01 – What building ClawRunr taught us about working with AI
- 00:38:31 – Closing
Links
- ClawRunr website
- ClawRunr on GitHub
- ClawRunr blog post
- JobRunr website
- JobRunr on GitHub
- JobRunr YouTube channel
- Spring AI