Inspiration
We noticed that finding creative collaborators in the streaming space is incredibly broken. Streamers post "looking for editor" tweets that get hundreds of replies, but there's no good way to evaluate fit. Meanwhile, talented editors get overlooked because they don't have 10K followers on Twitter. We wanted to build a system where work quality matters more than clout.
What it does
CreatorSync matches Twitch streamers with video editors based on creative compatibility—not follower count, appearance, or popularity. Editors upload sample clips anonymously, and AI suggests the best matches to creators looking for collaborators.
How we built it
Frontend:
HTML5, CSS3, JavaScript (ES6+) Lucide Icons (UI elements) Backend:
Node.js + Express.js (API server) SQLite (database) CORS (cross-origin support) AI Integration:
Aedify AI (text embeddings & semantic matching) Cosine similarity algorithm (matching scores) Tools:
Git/GitHub (version control) VS Code (development) Postman (API testing)
Challenges we ran into
- Designing the Anonymous Identity System
We needed to hide all identifying information until BOTH parties agreed Solution: Multi-stage data reveal (clip → bio → full identity)
- Integrating Aedify AI
First time using embeddings and semantic matching Solution: Built a fallback tag-based system that works even if AI fails
- Building a Fair Matching Algorithm
How do we pick 1 clip from an editor's 3-5 samples? Solution: AI ranks clips by similarity to creator preferences
- Backend-Frontend Connection
CORS errors and API configuration issues Solution: Proper middleware setup and environment variables
- SQLite Database Design
Modeling the two-sided matching flow in SQL Solution: Status-based match lifecycle (liked → accepted → confirmed)
Accomplishments that we're proud of
CreatorSync was a huge milestone for our team because most of us built our first full-stack application through this project. We’re proud that we didn’t just design a UI—we connected a real backend, set up routes, stored data, and made the frontend actually talk to the server. We also implemented role-based experiences (streamer vs. editor), built the core matching flow (browse → request → review → match), and got the project structured in a way that’s deployment-ready for Aedify. Overall, we’re proud we shipped a working end-to-end prototype that proves the concept and gives us a strong foundation to improve the matching logic and scoring system next.
What we learned
We learned multiple frameworks like Node.js, and how to implement these APIs. Another thing that we learned together was using HTML and JavaScript.
What's next for CreatorSync
Next for Creator Sync is scaling from a nightlife tool into a cross-industry platform by standardizing our data, expanding API access, and turning creator performance into actionable insights. That foundation lets us grow awareness and apply the model to spaces like arts, theater, and live culture—anywhere creators can drive real-world engagement.
Log in or sign up for Devpost to join the conversation.