binder.

"For You" page
Swiping functionality
Watchlist
Expanded description
Tech stack

Inspiration

We wanted to make financially conscious and sustainable shopping effortless. Buying used items saves money and reduces waste, but sifting through endless listings on Kijiji can be time-consuming and BORING! That's why we envisioned a "Tinder for Kijiji", an app that's financially conscious, fun, and eco-friendly.

What it does

binder. helps users search for and discover local secondhand deals with a simple swipe interface. It uses AI to estimate fair market prices, helping buyers make informed, sustainable financial decisions while promoting a circular economy.

How we built it

We combined Next.js and TypeScript for the frontend with Python (Selenium) for web scraping. Our custom scraper collects real-time data from Kijiji listings including title, price, description, and images and stores it in CSV files. The backend processes this data using Gemini Flash to determine fair market values, while the frontend presents the listings in a swipe-based “Tinder” interface.

We also integrated multiple data pipelines between the Python backend and the Next.js frontend to ensure consistent and fast data updates.

Challenges we ran into

One of the biggest challenges was connecting the frontend (Next.js + TypeScript) to the Python Selenium backend. Managing communication between these two different environments with live scraping and asynchronous data fetching was a significant hurdle.

We also faced difficulties in storing and fetching scraped data from HTML into our program efficiently, as well as keeping the interface responsive while processing listings in real time.

Accomplishments that we're proud of

We’re proud that we built our own scraping algorithm from scratch and successfully connected multiple programming environments together, getting Selenium, Next.js, and TypeScript to work in sync through many technical challenges.

What we learned

We learned that backend development is far more complex than it seems, and that AI tools alone can’t magically solve integration or communication issues between systems. Debugging, managing concurrency, and maintaining reliable data pipelines all require a deep understanding of both sides of the stack.

What's next for binder.

We plan to expand binder. with user profiles, and better AI models not only for condition-based price estimation, but for real-time tips on bargaining with sellers. We also wish to scale to other digital marketplaces and promote the circular economy on the national scale.