I've shipped a handful of side projects over the years. Most of them went nowhere — a few users, maybe some interesting feedback, then quiet. And I keep doing it anyway.

The real reason

The honest answer is that I learn best by building. Reading documentation is fine, but nothing sticks until I've broken something and had to figure out why.

Each project forces you to make decisions end-to-end: the database schema, the UI layout, the deploy pipeline, what to cut. That whole-system thinking is hard to practice in a day job where you often own just a slice.

What I've shipped

• Oisko.app — a social calendar tool. Taught me Supabase and real-time subscriptions.
• AdventureHub — an AI-assisted RPG campaign tool. Taught me how to work with the OpenAI API and stream responses.
• Reitit — a route planner using public transport APIs. Taught me to deal with external APIs that have bad documentation.

What goes wrong

The graveyard is bigger than the portfolio. Projects stall when the interesting technical problem is solved but the boring product work remains. I've gotten better at recognizing the moment when I'm building to learn versus building to ship — and being honest about which one I'm doing.

The takeaway

Ship something small. Then ship it again. The portfolio matters less than the habit.