Can You Really Build an App in 48 Hours Using AI Tools?
I usually write here about marketing analytics — dashboards, GA4, BigQuery.
But today’s story is different. It’s about my first app that we built using AI tools. And yes, we built it in just two days.
Last weekend, I joined an AI Hackathon in Lisbon. When I signed up, I honestly doubted that anything meaningful could be built in just two days. But I decided to give it a try with two teammates — and it turned into a great experience.
From Hypothesis to Prototype
We started with a simple hypothesis: people struggle to price second-hand items, and resellers often waste hours checking listings manually. Our idea was to create Sellis.pro — a tool that gives you an instant price estimate from just one photo.
Customer Research (CastDev)
To validate the idea, we first ran a couple of customer interviews (castdevs). The key insight: the biggest value is not for buyers, but for sellers — resellers on platforms like eBay or local marketplaces. Based on this, we shifted the focus of the product from “for everyone” to “for sellers only.”
Here’s what they told us:
- “I usually just scroll eBay listings for hours to guess a price.”
- “If I had something faster, I’d list more items.”
- “For buyers it’s interesting, but for sellers it’s money.”
That last one hit us: sellers get direct financial benefit, so we shifted focus from both sides of the market → to only resellers.
What We Built in 48 Hours Using AI Tools
We set up a quick landing page, launched a couple of promo videos, and even tested ads on YouTube to reach reseller communities. At the same time, we built the first working version of the product: upload a photo → get an estimated price.
The tools we used:
- Replit — to build the website
- Brevo — to collect leads via the landing page
- Cursor — to build and host the application
The Hackathon Experience
The event itself was huge: more than 1,000 people registered, and around 585 joined online and 260 offline. By the end, 130 projects were submitted by 323 participants.
We were proud to be among the teams that shipped — and even more exciting, our project was selected into the Top 15 best projects. Now we’re waiting for the final results and to see who takes the prize.
Takeaways
For me, the biggest value wasn’t just about the app — it was about the process:
- Working in a team offline, when you can literally turn to your teammate and solve problems in person.
- Pushing through deadlines and actually shipping something.
- Trying out new tools (I even built a website in Replit for the first time).
Whether or not we win, I already count this as a success. We built, we shipped, and we learned a lot in the process.
Product links
Check out the demo here: sellis.pro
Watch the promo videos: Video 1, Video 2