Building on the side of a full-time job

I write software for a living. Then I close the work laptop, open my own, and keep going.

That sounds like a lot. It isn't, if you're honest about how little time you actually have. Right now that time goes into Castfy, which I'm building in public. A few focused hours a week is the real budget. Everything else follows from that.

Scope is the whole game

With a day job, I can't out-work anyone. So I out-cut them.

Every feature has to answer one question: does this get the product in front of someone sooner? If not, it waits. Settings pages wait. Admin dashboards wait. "It would be nice if" waits forever.

Small sessions, finished things

I try to end every session with something done. Not "made progress on". Done. A fixed bug, a merged branch, a screen that works.

An hour that ends with something shipped beats three hours that end with a half-finished refactor I have to remember tomorrow.

Pick boring tools

The day job already gives me enough new problems. On the side, I use what I know: Next.js, React, Node, Postgres. Boring tools mean my limited time goes into the product, not into learning the stack.

Ship before it feels ready

I post demos before things are polished and ask people straight up: is this ready for beta? The answers are more useful than another week of polishing alone.

Solo is a feature

No meetings. No waiting for approval. I decide in the morning and ship at night.

The cost is that nobody catches my mistakes. So I keep things simple enough that I can catch them myself.

That's the whole system. Small scope, finished sessions, boring tools. It's slow, but it compounds.