Team 107_JobSwipe

Inspiration 😵‍💫

Job hunting as a fresh grad is messy and exhausting—you’re bouncing between job boards, spreadsheets, and random course links while trying to stay motivated. We wanted JobSwipe to feel like a single, guided flow that turns “I’m not sure what to do next” into clear next steps.

What it does 🚀

JobSwipe is an AI-powered career copilot that helps you go from browsing to offers faster:

Swipe-based Job Discovery to quickly shortlist roles
Compatibility Scoring so you can focus on the best-fit jobs first
Saved Jobs Hub to organize your shortlist in one place
Application Tracking across stages (Applied → Interviewing → Offered/Rejected) so nothing falls through the cracks
AI-Powered Multi-Step Career Roadmap tailored to your target role, broken into actionable tasks
Progress Dashboard that visualizes what you’ve completed and what’s coming next

How we built it 🛠️

Frontend (Modern Web App)
Built with Next.js + React + TypeScript, using a responsive, high-density UI with smooth interactions (including swipe-style browsing) powered by Framer Motion.
Backend (API + Data Layer)
Built with Node.js + Express to provide endpoints for listings, saved/applied tracking, user data, and roadmaps, with persistence in Supabase (Postgres).
Backend Architecture (MVC)
Organized using Model–View–Controller: controllers handle request/response logic, while models manage database operations and AI-related work—keeping concerns separated and iteration fast. Ensures scalability and ease of maintenance.
AI Layer
Integration with OpenAI model to generate structured roadmaps containing actionable tasks for the users to use as a checklist/progress system.

Challenges we ran into ⚡

Keeping AI output consistent required careful prompting and validation.
Performance tradeoffs when scoring lots of jobs—balancing responsiveness with useful ranking.
State consistency across views (saved jobs, applications, roadmap progress) so updates feel instant and accurate.

Accomplishments that we're proud of ✅

Built an end-to-end flow: discover → score → save → apply → track → upskill in one app in 24 hours.
Turned AI into something actionable (structured roadmaps), not just “chat output”.
A dashboard experience that makes progress visible and motivating day-to-day.

What we learned 📚

A lot can be done in 24 hours if you put your mind to it.
AI features are only “good” when they’re reliable and productized—schemas, fallbacks, and UX matter as much as the model.
Progress tracking creates momentum—people stick with plans when progress is obvious.

What's next for JobSwipe 🗓️

Persist roadmap completion across devices so progress is never lost.
AI-powered skill-gap insights that automatically recommend resources tied to each roadmap step.
Reminders + scheduling for interviews, follow-ups, and learning sessions.
Integration of tools to help improve Resume and Interview Skills.

Swipe today, celebrate tomorrow!