Design changes are simple, quick, and cheap. Engineering changes are difficult, lengthy, and costly. That’s why, when it comes to complex systems, the rule is: design first, build second. Multi-agent systems (MAS) are such complex systems: agentic AI systems consist of multiple AI agents and—like any socio-technical system—one or more human agents (i.e., their users).
Whenever you’re faced with complexity, you won’t get very far on your own. Think together to design better systems.
At Datentreiber, we’ve developed a tool specifically for this kind of collaborative Design Thinking workshop. The Multi-Agent System (MAS) Design Workshop is one of our most popular free templates on Miroverse.
There’s now also a video from the Data Science Summit 2026 featuring a half-hour introduction:
👉 https://youtu.be/Lue4LecPbdM
We’d love to hear your questions and feedback in the comments! 👇 If you like what we do, give the video a 👍 on YouTube and the template a ♥️ on Miroverse. Thank you! 🙏
