What I care about when building products
No shortcuts on understanding users, planning, design, or testing
AI can now take on the implementation. That's why I do the rest myself and don't compromise on it: watching people use things to find the real problem, planning what to build and in what order, designing the screens and flows, and testing what I made.
The everyday hassle first, then the tech
I start from small hassles that I and the people around me run into every day, not from interesting technology. For me, I think that's the only place really used things come from.
Will people open it, not just does it work
No matter how capable something is, it means nothing if people don't bother to open it. Don't make them open something new, don't make them type. Lowering the hurdle comes first.
Stick to the problem, not the method
I try not to get attached to a technology or an approach. What matters most is reaching the problem, so if the method I chose isn't accepted, I look for another way.
Speed is a weapon, not the goal
AI lets me build very fast. Even so, I want to keep the option of taking it slow. I try not to be pushed around by speed itself.
AI builds it, so I guard it twice as hard
Because I hand implementation to AI, I pay extra attention to security, data leaks and corruption, and keeping things running for a long time. Holding users' data is my responsibility.
Writing
More pages