I'm not technical, but my friends and coworkers are. We always have discussed plans to build large projects and video games. I have no technical experience, but they encouraged me to try building something with AI and to start small.
I picked a problem I had + a product I knew I would use myself. I've used Pomodoro timers on and off for years to track my focus time. I fell off routinely. Then I realized how long I'd happily grind something monotonous in a game just for the chance at a shiny or a rare drop. So I made a timer where every finished focus session spawns a little pixel monster you throw a tomato at to catch, with a small chance it's a rare one. I called it pomomons.io.
I built it with Claude. I made the decisions and tested everything. A few things I learned:
I commonly assumed things couldn't be done with Claude that could. I spent a couple of hours building a promo video only to be trashed by the one Claude made in a couple minutes.
Don't believe something is fixed until you see it. My logo got "fixed" three times with no change on my screen, because it was being tested on a different screen size than mine.
The weird bugs happen when you're not looking. You have to use the app like a real person would.
Write your plans down somewhere permanent. Mine lived in one chat for weeks and would have been gone if it got cleared.
Four months later it's live at pomomons.io. It's free, works offline, and doesn't need an account. And it works on me. Wanting to see what shows up at the end of a session is often what gets me to start the next one.
If people keep telling you to build something, pick a problem you actually have and start small. You'll care enough to keep testing it, and that matters more than knowing how to code.
Give it a try and give me feedback pomomons.io
Top comments (1)
Pomomons is a great name for this. It tells me Pomodoro plus monsters in one word, so I understood the idea before I even opened the site. The rare monster chance is the clever part too, it gives people a reason to finish one more session. For a first build, starting with a problem you had yourself was the right call.