Iotopia: Bridging Hardware and Software
My second hackathon started with confidence the first one never had. I'd spent months as a product development intern at ArkVerse by that point. Built real things for real users, shipped them, learned from the wreckage. I knew how to approach a 24-hour sprint. Scope aggressively. Build fast. Don't waste time on anything that won't show up in the demo.
My second hackathon started with confidence the first one never had. I'd spent months as a product development intern at ArkVerse by that point. Built real things for real users, shipped them, learned from the wreckage. I knew how to approach a 24-hour sprint. Scope aggressively. Build fast. Don't waste time on anything that won't show up in the demo.
Then the organizers threw a curveball.
Same problem statement as my first hackathon: a platform for sharing medical records. But with a mandatory twist. Blockchain integration. The data had to live on-chain, tamper-proof and verifiable. Conceptually sound. Practically terrifying, given that not one of us had ever touched blockchain development.
We built on what we knew and bolted on what we didn't. Next.js and Firebase for the familiar parts. MetaMask for wallet integration. Solidity for smart contracts, which we were learning in real time as we wrote them. The first few hours went well. UI came together fast. Data models were clean. Auth actually worked on the first try, which felt like a small miracle after the Supabase nightmares at FSDC.
Blockchain broke that momentum completely.
Connecting a frontend to a smart contract isn't something you learn in an afternoon, but that's what we were attempting. Transaction handling was confusing. Gas fees introduced a cost layer nobody had planned for. Confirmation times destroyed any illusion of responsive UX. A user clicking a button and staring at nothing for fifteen seconds while a transaction confirms doesn't feel like a feature. It feels broken. We spent hours building loading states and confirmation flows just to make the waiting feel intentional rather than bugged.
Every step deeper surfaced two more concepts we didn't understand. Wallet connection states. Nonce management. Contract deployment quirks. Event listener patterns. Learning-by-drowning, which is the best way to learn if you don't mind the panic.
We had a working prototype by the deadline. Medical reports stored on-chain, accessible through a clean interface, verifiably tamper-proof. Not production-ready by any stretch, but enough to prove the concept worked.
What stuck wasn't the code. It was watching other teams solve the same problems with entirely different approaches. Some had actual Solidity experience and were building things we couldn't even conceptualize yet. Others were stuck on the same wallet integration bugs we'd burned hours on. The gap between "knows blockchain" and "learning blockchain at a hackathon" is enormous, and being on the wrong side of it for 24 hours gave me respect I didn't have before.
Zero hours of sleep. A functional prototype. And a genuine appreciation for anyone writing smart contracts full-time. The tooling is rough, the abstractions leak constantly, and the feedback loops are painfully slow compared to standard web development.
No demo video for this one, but the FSDC demo shows the core concept well enough. Same medical records platform, same idea. Just add MetaMask popups and gas fee anxiety.