I have a strong hypothesis about what’s coming next, and I’m already testing it with Anapana. This app and the next ones will be my living lab to validate it in the real world.
Hope we can dive into it soon over a coffee, even a virtual one ☕️
This is a masterclass in what high-agency building actually looks like in the AI era.
What stood out most isn’t the speed. It’s the discipline. You didn’t “vibe build” your way to a demo. You treated AI like a precision instrument by doing the hard, unglamorous work upfront: context, constraints, pedagogy, edge cases, and success criteria. That’s the difference between AI as a toy and AI as a force multiplier.
The ChromaKid example lands especially well because it grounds the hype in real cognitive science and real users with real stakes. Toddlers don’t tolerate jank, and neither do parents. Designing for that constraint forces clarity in a way most MVPs conveniently avoid.
There’s also a quiet but important lesson here for solo builders and creators alike: speed doesn’t replace thinking, it exposes it. When iteration collapses from weeks to minutes, fuzzy thinking has nowhere to hide. Planning becomes the product.
This feels less like “look what AI can do” and more like “look what becomes possible when clarity meets courage.” That’s the part worth emulating.
Thank you so much for this comment, you really nailed the point. That’s exactly the trick.
I applied the same approach to build a mobile app (it's called Coco), and it’s honestly impressive what one person can do with solid thinking and clear constraints. It’s amazing what people can build today.
Amazing article Marco! Your experience in product management and development really makes a difference in the analysis and prompting part, and I think that's the big difference between using AI with a plan or throwing prompts at it and hoping for the best!
That's a great example... Clearly now the challenge is vibe go to market to achieve this kind of performance when you are selling the product :)
100% agree.
I have a strong hypothesis about what’s coming next, and I’m already testing it with Anapana. This app and the next ones will be my living lab to validate it in the real world.
Hope we can dive into it soon over a coffee, even a virtual one ☕️
Come sempre illuminante !!! Grande Marc !
Grazie mille Michele! 🙏 Fa davvero piacere leggere commenti così. A prestissimo!
I think the icing on this cake would be to actually see this happen. Is there a video we can watch along the way?
Not yet, but I’m working on something 🙂 Stay tuned.
This is a masterclass in what high-agency building actually looks like in the AI era.
What stood out most isn’t the speed. It’s the discipline. You didn’t “vibe build” your way to a demo. You treated AI like a precision instrument by doing the hard, unglamorous work upfront: context, constraints, pedagogy, edge cases, and success criteria. That’s the difference between AI as a toy and AI as a force multiplier.
The ChromaKid example lands especially well because it grounds the hype in real cognitive science and real users with real stakes. Toddlers don’t tolerate jank, and neither do parents. Designing for that constraint forces clarity in a way most MVPs conveniently avoid.
There’s also a quiet but important lesson here for solo builders and creators alike: speed doesn’t replace thinking, it exposes it. When iteration collapses from weeks to minutes, fuzzy thinking has nowhere to hide. Planning becomes the product.
This feels less like “look what AI can do” and more like “look what becomes possible when clarity meets courage.” That’s the part worth emulating.
Thank you so much for this comment, you really nailed the point. That’s exactly the trick.
I applied the same approach to build a mobile app (it's called Coco), and it’s honestly impressive what one person can do with solid thinking and clear constraints. It’s amazing what people can build today.
I didn’t know you played piano in a jazz band 🎶 by the way, congrats 🎉
Yesss! 🎶 Hope to jam sooner or later!
Great article! I enjoyed it, as a musician and as AI daily user.
Thank you so much! 🙏
Amazing article Marco! Your experience in product management and development really makes a difference in the analysis and prompting part, and I think that's the big difference between using AI with a plan or throwing prompts at it and hoping for the best!
Thank you! I totally agree.
I’m lucky to have these skills, so in some ways I start a bit ahead... but AI is really democratizing product building. The gap is shrinking fast.
Your perfect pitch IS a superpower! Especially since there are no musicians in your family. That's fascinating.
Thank you so much! 😊
Great guide, Marco! This is exactly the right approach, at least it has worked for me. Plan ahead before using Lovable, or it’ll consume your credits!
Thank you Elena! Planning is essential.
I’m going to release soon a complete post on how I developed the production app... its name will be Coco 🚀