An AI-powered course and club planner for UC San Diego students.
Currently in active development (MVP sprint: 17 days).
This project is in progress as part of a 17-day MVP build plan.
The goal: ship a working full-stack prototype by the end of the sprint.
Current progress:
- ✅ Next.js + TypeScript frontend scaffold
- 🚧 PostgreSQL + Prisma backend setup
- 🚧 AI integration with GPT-4o + LangChain
- 🚧 Clerk authentication & user profiles
- 🚧 Docker + AWS deployment pipeline
Daily progress is tracked in docs/plan.md.
- Course Scheduling – drag-and-drop UCSD course planner
- Club Discovery – recommendations tailored to user schedules
- Smart AI Suggestions – GPT-4o powered academic/club balance advice
- Secure Accounts – login and persistence with Clerk + PostgreSQL
- Cloud Deployment – containerized with Docker, running on Vercel + AWS
Frontend: Next.js (App Router), React, TypeScript, TailwindCSS
Backend: FastAPI + Node.js API routes
Database: PostgreSQL + Prisma ORM
AI/ML: OpenAI GPT-4o, LangChain
Infrastructure: Docker, AWS (S3/EC2), Vercel
- Node.js (v18+)
- PostgreSQL (local or Dockerized)
git clone https://github.com/jumanzor/triton-pathways.git
cd triton-pathways
npm install
npm run devOpen http://localhost:3000 to view the app.
Copy .env.example → .env.local and configure:
DATABASE_URL=...
OPENAI_API_KEY=...
CLERK_SECRET_KEY=...This README will expand as milestones are completed.
For now, Triton Pathways is a work-in-progress MVP sprint, targeted for completion in 17 days.
Stay tuned for updates as the project evolves!