RENEW'D

RENEW'D Login Page (Figma)
Add a Listing 1
Add a Listing 2
Add a Listing 3
Feed 1
Feed 2
Send Offer 1
Send Offer 2
Likes 1
Likes 2
Matches
Profile 1
Profile 2

Inspiration

Fast fashion is killing the planet. Millions of tons of textiles hit landfills every year while perfectly good clothes sit unworn in closets. Buying new is expensive, and existing thrift platforms are overwhelming catalogs with zero personalization. We saw an opportunity to turn your closet into someone else's favorite piece of clothing, and make sustainable fashion actually feel good. Why should finding your next favorite piece be harder than finding your next date?

What it does

Renew'd reimagines thrifting as an effortless, personalized experience. Users swipe through items matched to their style, sizing, and location—no cash, no shipping costs, just direct trades. Our matching algorithm pairs people based on what they're looking for and what they have to offer, keeping fashion circular while reducing waste and expanding wardrobes. Like something? Match with the owner, chat in-app, and arrange a trade. A world of preloved items now at your fingertips.

What makes us unique:

Matching algorithm: dating-app meets marketplace
Cash free: item-for-item trading, no shipping costs
Smart results: pairing users based on sizing, style and location

How we built it

We started by mapping complete user flows from sign-up through listing creation, swiping, matching, and trade completion. The database architecture uses relational structures connecting users, listings, likes, matches, and messages with optimized queries for real-time feed generation. The frontend centers on gesture-based swipe mechanics with smooth card animations and an intuitive, accessible UI with readable layouts and size-inclusive tags. The backend implements a weighted scoring system that normalizes location distance, size compatibility, and style preferences—both explicit (tags, requirements) and implicit (swipe history, interaction patterns). We designed the system around mutual matching rather than buyer-seller dynamics, which fundamentally changed how we approached transaction flows and incentive structures.

Challenges we ran into

Balancing algorithm feature weights across heterogeneous data types was complex—location needs distance-based scoring, style requires categorical matching, sizing demands precision. We had to optimize database queries joining multiple tables while filtering already-seen items and maintaining performance. The biggest challenge was rethinking marketplace dynamics for trades instead of purchases, which meant designing for mutual interest and equal exchange rather than one-directional transactions. Time constraints forced us to prioritize the core matching pipeline over features like trade histories, reputation systems, and community tabs.

Accomplishments that we're proud of

We successfully translated dating app engagement mechanics into a functional trading marketplace that solves real environmental problems. The database schema we built scales efficiently despite complex many-to-many relationships across users, items, and interactions. Our matching algorithm framework uses collaborative filtering techniques that improve recommendations over time as it learns from user behavior. We created comprehensive user flows covering every edge case and state transition. Most importantly, we proved that making sustainability effortless and fun can drive meaningful behavioral change without sacrificing functionality.

What we learned

We gained hands-on experience with recommendation systems, feature engineering, and multi-dimensional scoring algorithms. The project reinforced how critical database indexing and query optimization are for apps with complex relational data. We learned to adapt UX patterns between domains while preserving their psychological effectiveness—channeling the dopamine hit of swiping toward sustainable behavior. Building for trades rather than sales taught us to think differently about incentive structures, mutual value creation, and transaction completion. We also learned that accessible design and inclusive features (like size-inclusive tags) aren't just nice-to-haves—they're essential for building platforms that serve diverse communities.

What's next for RENEW'D

We're completing the MVP and launching a beta to gather real user data for algorithm refinement. Next comes machine learning integration；collaborative filtering for better recommendations and computer vision for automated item tagging and style recognition. We're building features like profile viewing, bundling multiple items in trades, like history tracking, and community tabs. The technical roadmap includes horizontal scaling with caching layers, materialized views for feed generation, and message queues for real-time notifications. We'll implement trust and safety features including reputation scoring and trade verification. Long-term, this matching framework could extend beyond fashion to furniture, electronics, books—any marketplace where discovery beats search and sustainability matters. We're not just reducing waste; we're making circular fashion the path of least resistance.