Înapoi la Perspective

Perspectivă din 30 octombrie 2026

English

I tend to talk a lot when I use voice typing (and in general, my friends would say). More context, more explanations, more details. In my mind, that should lead to a better result. At Shopware Hacktoberfest, Martin Bens' talk gave me another question to ask: what should the AI agent leave alone?

Sander and I discussed this in The Open Podcast episode about the event. In this clip, I talk about defining non-goals and setting clear boundaries.

What are we trying to achieve? What is outside the scope? Where should the work stop?

These are familiar questions in operations and project management. They are just as useful when we work with AI. We can describe the goal in detail and still leave too much room for assumptions, extra work or changes we never asked for.

Thank you to Shopware for bringing us together at Hacktoberfest, to Martin for the session, and to Sander for the conversation.

Full episode in the comments ⬇️

Citește subtitrările

This was very interesting for me because now with voice typing I tend to yap and talk and talk a lot and give a lot of context because in my mind it is that more context means better output. However, it was very interesting how how Martin came in and said, How about defining the non-goals? How about looking at what they don't need to do and make sure that your scope is clear and you have clear boundaries?