The age of personal software

When I was a student, I had something I didn't appreciate at the time: lots of free hours.

So I built things. Small things, mostly for myself. A script here, a little web app there, anything that made a boring part of student life a bit easier. When a tool worked for me, I shared it with friends. When it worked for my friends, it sometimes spread to the whole school.

One of them still makes me smile. Every exam season, my university published the exam schedule as hundreds of Excel files. To find your own exams, you first had to look up which files matched your class codes, somewhere else entirely. Then you opened each file and searched for your student ID, one by one. It was boring, slow, and easy to get wrong. And sometimes the school posted updates, more than once, so you had to do it all again.

So I built a tiny web app. You typed in your student ID, and it built your personal exam schedule instantly. The README even said the time you saved could be spent going on a date.

The code was small and simple. For a day or two, almost every student on campus used it.

That feeling never left me. Small code, big impact.

Then life got busy

After school, I did what most people do. I started working. The work was bigger, the problems were harder, and the days were full.

Side projects quietly disappeared. Not because the ideas stopped coming. I still noticed small annoying things everywhere, little problems that a weekend of code could fix. I just didn't have the weekends anymore. A tool that might help fifty people never felt important enough to beat a deadline, a meeting, or sleep.

I think a lot of builders know this feeling. The ideas pile up in a notes app and stay there.

AI gave me my weekends back

Something has changed in the last couple of years. With AI, the gap between "this would be nice" and "this exists" has shrunk dramatically.

The small tool that used to take me a few weekends can now take an evening. I describe the idea, AI helps with the boring parts, and I spend my energy on what matters: understanding the problem and making it feel right for the people who'll use it.

It's not magic. Building something people can rely on is still simple, but not easy. But the cost of trying is lower than it has ever been.

So I've started doing it again. Building small things. Repeating the pattern from my student days: solve it for myself, share it with a few people, and see if it helps someone else.

Software for a few people

I believe this is bigger than me. We're heading into a time of personal software.

For decades, software had to serve millions of people to be worth building. Small problems for small groups were never worth the cost. Now they are. A teacher can make a tool just for his class. A shop owner can build exactly what his team needs. A student can fix something annoying for his whole campus, in an evening.

Not every tool needs to be a startup. Some just need to make a few people's day a little better.

That's what I loved about building when I was young. And for the first time in years, I have the time to do it again.

What would you build if it only took an evening?