You're not doing a new AI model. It's more about how do I use the APIs and the frameworks and the tech stack that we have? How do I put that together to build something? Then if the problem is clear and the time is now, you will be able to come to a conclusion about what's possible to build in three sessions, something like that. The key thing is leaning into those sessions and really wrestling with each other. If you have just the product and the designer there and then it's like, well, we'll show this to the technical person later, then it can all blow up. And then you find out it's more complicated than you thought and you have to go back to the drawing board. We need to have all the necessary information in the same room for these sessions to go fast.
You're not doing a new AI model.