A full-stack transit and ride-sharing platform built for students, staff and faculty at Purdue University.

Students at Purdue University face a persistent problem: getting between the West Lafayette campus, the Indianapolis campus, and the major regional airports (IND/ORD). This is due to it being expensive, unreliable, or both. Existing solutions are fragmented, there's no single, student-first platform that handles both free on-campus shuttles and low-cost regional rides.

Boiler Rides solves this with a unified, Boiler-branded web application where students can:

• Book free on-campus shuttle routes using their Purdue ID
• Reserve low-cost regional rides to Indianapolis, O'Hare, and surrounding hubs
• Browse interactive maps to find and visualize nearby pickup points
• Manage their account and booking history in a secure, student-only environment

The platform bridges the gap between campus life and regional mobility - all under one roof.

Our team moved from concept to a fully functional MVP during the hackathon. Here's a breakdown of every core module shipped:

The heart of the platform — a robust seat-booking system designed to handle concurrent requests safely.

• Atomic Transactions: Seat reservations use MongoDB's findOneAndUpdate to atomically increment seatsBooked only when seatsBooked < totalSeats. This prevents race conditions and overbooking without the need for distributed locks.
• Multi-Step Booking Funnel: A 5-stage checkout process — Date → Route → Time → Details → Checkout — that breaks complex transit data into a digestible, linear user flow. Each step validates before progressing, reducing errors and drop-offs.

Students shouldn't have to guess where to go. The platform uses real location data to surface the most relevant pickup points.

• Real-Time Proximity Calculation: Implements the Haversine Formula to compute the precise distance (in miles) between the student's current geolocation and every available pickup hub.
• Interactive Mapping: Leaflet markers with "FlyTo" animation logic let users visually inspect pickup locations at airports and campus landmarks. Clicking a hub smoothly pans and zooms the map to that location.

The platform is designed to be a trusted, closed environment for Purdue students only.

• Verified Registration: The signup flow enforces .edu email domain validation, ensuring only students can create accounts and keeping the community safe.
• JWT Authentication: Stateless token-based sessions keep the server lean and responsive without storing session state.
• GDPR-Compliant Account Deletion ("Right to be Forgotten"): When a user deletes their account, the system cleanses all personal data and restores any booked seats back to the available inventory pool — maintaining data integrity across the platform.

To simulate a living transit network, the team built a dynamic data seeder.

• 14-Day Rolling Schedule: Generates 1,000+ unique ride instances covering the next two weeks, automatically refreshing to always reflect a realistic upcoming timetable.
• Tiered Pricing: Rides are categorized as Free (on-campus shuttles) or Paid (regional routes), with pricing logic baked into the seeder.
• Symmetric Duration Matrix: Arrival times are computed dynamically from a route duration matrix where every route pair is consistent in both directions (e.g., PMU ↔ IND is always 105 minutes). This ensures schedule integrity without hardcoding individual times.

1. Atomic Overbooking Prevention — MongoDB's findOneAndUpdate with a conditional guard handles concurrent booking requests at the database level, not the application level.

2. Haversine Geolocation — Distance to pickup hubs is calculated on the fly using the Haversine Formula, with no third-party geolocation service dependency.

3. Stateless Architecture — JWT-based auth means the Express server holds zero session state, making it horizontally scalable from day one.

4. Symmetric Route Matrix — The duration matrix ensures bidirectional route consistency, eliminating the need to maintain separate outbound/return schedules.

5. UI/UX Polish — Map fly-to animations, card scaling effects, and confetti on booking completion elevate the experience well beyond a standard form-based app.

• Node.js v18+
• MongoDB (local or Atlas)
• npm or yarn

# Clone the repository
git clone https://github.com/your-org/boiler-rides.git
cd boiler-rides
 
# Install dependencies for both frontend and backend
npm install

# Set up environment variables
cp .env.example .env
# → Fill in MONGO_URI, JWT_SECRET, and PORT
 
# Seed the rolling schedule (generates 14 days of rides)
npm run seed
 
# Start both servers concurrently
npm run dev

The frontend will be available at http://localhost:5173 and the API at http://localhost:3000.

Boiler Rides was built with AI assistance throughout the development process:

• Claude — Used to craft a detailed prompt based on the features we wanted, which served as the foundation for code generation.
• Amazon Q — Took the Claude-generated prompt and scaffolded the initial BoilerRides application.
• Gemini — Used in the finalization phase to debug and clean up frontend and backend issues.

Built with ❤️ at HackIndy by the Boiler Rides team - Ramitha Viswasekar, Mridini Kullkarni, and Meraj Syeda