As students, we realized that working on projects is one of the fastest ways to learn — yet finding the right teammates is always the hardest part. People struggle with:
Mismatched skills Unavailable teammates Random team assignments No clear idea of what they need to learn next
We asked ourselves: “What if forming teams could be as smart as using AI, and as fun as taking a quiz?”
That question inspired SkillSync AI — a platform that uses AI to automate learning, build smarter study groups, and help students grow together.
What We Learned
During the hackathon, we explored topics we had never used in a real project before:
React + TypeScript Tailwind CSS + multi-theme UI (Light, Dark, Pink mode) Designing AI-driven workflows Integrating GitHub skill analysis Authentication & state management Building a quiz-based matching engine Recommender systems for personalised learning
We especially learned how to design systems that guide users to grow, not just match them randomly.
How We Built It
SkillSync AI was built using a modern full-stack workflow:
Frontend React + TypeScript
Figma → for design
3 Themes: Light (Blue), Dark (Purple), Pink (Fun Mode)
Interactive pages: Dashboard, Quiz, Gap Finder, Team Assignment
AI Features
Gap Finder AI: Takes project details and identifies missing features, skill gaps, and role optimization.
Quiz-Based Team Formation: Matches users using vector scoring:
Match Score
𝛼 𝑆 + 𝛽 𝑃 + 𝛾 𝐴 Match Score=αS+βP+γA
where S = skills, P = personality, A = availability.
Learning Recommendations AI: Suggests courses, project ideas, and GitHub contributions.
Backend
Node + Express API
GitHub profile analyser
Matching engine
MongoDB for user data
Discord Automation
When a project is created, our Discord bot auto-creates a private team channel using:
Channel
TeamName +
ProjectID Channel=TeamName+#ProjectID
Challenges We Faced
- Designing a Multi-Theme System
Figma generated hundreds of styles — implementing them manually caused conflicting CSS variables and broken dark mode. We rewrote the entire theme engine using:
[data-theme="dark"] { ... } [data-theme="pink"] { ... }
- Debugging a Broken UI Export
AI-generated components were inconsistent; many buttons didn't work, states were missing, and navigation failed. Fixing this taught us:
Component structuring
Prop drilling
Conditional rendering
- Matching Algorithm Complexity
Creating fair team assignments was hard. We finally built a weighted scoring system that balances:
Skill-fit
Personality compatibility
Availability
Project requirements
- Time Pressure
Integrating AI + UI + backend + Discord automation within hackathon time was extremely challenging — but also the most valuable learning experience.
Final Thoughts
SkillSync AI is more than a project — it’s our attempt to make learning collaborative, smart, and fun again. We built it because we believe students learn fastest when they learn together, and AI can make that process smoother, fairer, and more exciting.
drive link : https://drive.google.com/drive/folders/1Q9mYq84OGMnDDq2sQvG252AF1h5pHmQA?usp=drive_link
Built With
- css3
- dark
- express.js
- figma
- github
- mongodb
- node.js
- personality-&-availability)-ai-gap-finder-(project-analysis-&-missing-features-detection)-ai-recommendation-engine-(courses
- pink)-lucide-icons-figma-?-react-ui-export-backend-node.js-express.js-mongodb-/-mongoose-jwt-authentication-ai-&-algorithms-custom-matching-engine-(weighted-score-using-skills
- react
- replit
- tailwind
- typescript
- vercel
- vite
Log in or sign up for Devpost to join the conversation.