Flashdrop

Inspiration

Weekend sellers still rely on paper signs, cash, and messy manual setup. We wanted to make selling in person feel as fast as taking a photo and speaking one sentence.

What it does

FlashDrop turns a photo and voice pitch into a live product listing in seconds. It generates the title, description, price, inventory, and checkout flow, then uses bunq to power QR-based payments, live balance context, and real-time sale updates.

How we built it

We built FlashDrop with a React frontend and a FastAPI backend. We used multimodal AI for transcription and listing generation, bunq APIs for payment tabs, account balance, and webhooks, and server-sent events for live dashboard and wall updates.

Challenges we ran into

The biggest challenge was reliability across the full flow: getting AI output to feel fast and useful, handling bunq sandbox limitations, and making payment events update the UI instantly and consistently.

Accomplishments that we're proud of

We built a complete end-to-end experience, not just a concept. A seller can create a drop from voice and camera input, publish it, let buyers haggle, accept payment, and watch inventory and notifications update live.

What we learned

We learned that AI works best when it removes friction from an existing behavior instead of creating a new one. We also learned that real-time financial feedback is what makes the product feel alive rather than simulated.

What's next for Flashdrop

Next we want to harden the live bunq integration, improve real checkout reliability, support richer seller workflows, and make FlashDrop robust enough for actual student markets, booths, and pop-up events.