I thought I was cooked. Then I found an empty cardboard box.

Six hours into the hackathon, our team had nothing. Not "nothing good" — just nothing. We'd been throwing ideas at a whiteboard that weren't sticking, everyone half-convinced we'd already lost before we started. I was fully in my head. I stepped away to grab a snack.

On my way back, I walked past a corner of the room with a pile of empty cardboard boxes — the kind Amazon ships monitors in. I don't know why, but I stopped. Something about the shape of it. A slot at the top, a door at the front. This looks like a vending machine.

That's when it clicked.

We'd been trying to think of something "impressive" instead of something useful. The idea came fast after that: what if medication pickups worked like a vending machine? Pharmacies in low-income and rural areas are understaffed, often inaccessible, and the last mile of prescription delivery is genuinely broken. GUMDROP — a smart, Arduino-powered dispensary kiosk that optimizes convenience and security for underserved communities.

We built it in the next 18 hours. Won.

The thing I keep thinking about: the best ideas I've had didn't come from staring at a problem directly. They came from stepping back, letting your brain wander, and noticing something random that reframes everything. The cardboard box didn't give me the idea — it gave me permission to think differently.

Now whenever I'm stuck, I go get a snack.