Whenever I’m not building a product or tinkering with a side project, I keep trying to find things to write about.
I look at X, LinkedIn, blogs, product launches, new AI tools, and there is always a lot happening. There is no shortage of information.
The problem is not a lack of ideas. It is too many scattered ones, and none of them feel focused enough to turn into something useful.
The moment I start building, that changes. Every decision, mistake, tradeoff, and small discovery gives me something real to write about.
Which LLM actually works better for this use case? Is the more expensive one worth it? What was easy to integrate and what became a mess? Which feature sounded useful but failed in practice? What changed after testing it with real users?
These are not topics I have to invent. They come directly from the work.
That is why I like the idea that action produces information. Building forces you to make decisions, and those decisions give you something specific to learn from and write about.
You can read a lot and still not know what actually works. Once you start building, you get your own answers.
Since joining SpaceXAI, I have not had much time to build something of my own.
A while ago, I started ShipDemos.com but never really found the time to work on it properly. I have been thinking about picking it back up.
Building gives something real to think about, test, and eventually write about.
So I’m picking ShipDemos back up. I’ll share what I learn along the way.