Ever wanted to pick up a new hobby and had no idea where to even start? Not which one to try, and not what to do once you've picked one. That's the real problem Chloe spent her residency solving.

Meet Chloe

Chloe is a high school student in a Flintolabs residency, where you're handed a real problem and asked to build something that actually solves it. Her assignment: help people get past the blank-page freeze that hits the second they decide they want a new hobby but don't know where to begin.

The problem

Starting something new is harder than it sounds. There are a thousand hobbies out there, no clear way to know which one fits you, and even once you pick one, the first few weeks can feel like staring at an empty page. Most people just... don't start.

The brief was clear: build something that takes a person from "I have no idea what to try" to "here's exactly what I'm doing this week."

The build

That brief became a live web app. You write a short blurb about what you're looking for, and the app suggests three hobbies with a genuine reason each one might fit you. Pick one, and it hands you a checkable 4-week beginner plan, so the first month isn't a guessing game.

Chloe built this the way real software gets built: piece by piece, in the right order. First the suggestion engine, then the screens a user actually clicks through, then the 4-week plan generator. She's now in the middle of the hardest part, saving each plan to a database so it's still there the next time someone opens the app instead of disappearing on a refresh. That's genuine full-stack work, and she got there on her own timeline, moving steadily week over week.

What stands out isn't just that it works. It's how deliberately she got there. She iterates on her prompts the way an engineer iterates on code: "I started simple and built up until it gave the exact answer I wanted." She can explain why a particular design choice fits the user, not just that it does. And the database feature wasn't handed to her as a task, she noticed on her own that saved plans needed to persist, and started building it before anyone had to ask.

Why this matters

This is what we mean when we say build with AI, not just use it. Chloe didn't follow a tutorial on how to build a recommendation engine. She reasoned through her own prompts until they gave her exactly what she needed, spotted a missing feature no one flagged for her, and is now working through the genuinely hard part of shipping software: making data stick around. That kind of judgment doesn't come from watching someone else build. It comes from building.

Tell Chloe what you think

Chloe's hobby-matching app is live on the Flintolabs Marketplace. We release student-built tools like this one there because if Chloe ran into this problem, chances are plenty of other people have too, and this app might already solve something you're dealing with right now.

Describe what you're looking for, see what it suggests, and then leave Chloe feedback. Tell her what you love and what still needs work. That feedback loop, real users telling a real builder what to fix, is exactly the kind of experience that turns a class project into an actual skill.

You can also reach the app directly at hobby-match-mu.vercel.app.

Have you ever wanted to start something new and just didn't know where to begin? Tell us in the comments.