NYU Course Navigator 🎓

A production-grade semantic course search platform for NYU students (Feb 2026)

A robust, full-stack application to scrape, index, and semantically search 19,000+ NYU courses. Built with FastAPI, React 19, PostgreSQL (pgvector), and Redis.

---

⚡ Quick Start (Windows, Mac, Linux)

1. Clone the repository:

git clone https://github.com/harshagarwalnyu/nyu-course-search.git
cd nyu-course-search

2. Setup Environment Variables: Copy .env.example to .env and fill in your API keys (see API Keys section).

cp .env.example .env

3. Start the application:

# Option 1: Start Core Stack (Backend, Frontend, DB, Redis) - Recommended
make dev

# Option 2: Start Full Stack (Includes Neo4j & Workers)
make dev-full

Access the application:

• 🌐 Frontend: http://localhost:3005
• 📚 Backend API: http://localhost:8005/docs

---

🚀 Features

• 🧠 AI Semantic Search: Find courses by concept (e.g. "ethics in AI") using pgvector.
• 🔐 NYU Auth Wall: Strict student-only access via @nyu.edu OTP verification.
• 📊 Massive Index: Over 19,000 courses scraped from NYU Bulletin and Albert.
• 🎨 Modern Dashboard: Responsive UI with dark mode, real-time filtering, and "Thinking UI" traces.
• ⚡ Optimized Dev: Docker profiles for fast startup and uv for lightning-fast Python installs.

---

🔑 API Keys

To enable AI features, add these keys to your .env:

Service	Key Name	Purpose	Cost
Hugging Face	HF_INFERENCE_API_KEY	Vector Embeddings	Free
Jina AI	JINA_API_KEY	Search Reranking	Free Tier
Groq	GROQ_API_KEY	Llama 4 Reasoning	Free Beta
OpenRouter	OPENROUTER_API_KEY	Graph Extraction	Cheap

---

🛠️ Available Commands

Command	Action
make setup	Initial build and dependency pull
make dev	Start Core stack in background (Fast)
make dev-full	Start All services (Neo4j, Temporal, Workers)
make logs-core	View live logs for the core application
make build	Rebuild all containers in parallel
make clean	Stop and wipe all data/containers
make migrate-embeddings	Regenerate all 19k course vectors

---

🏗️ Architecture

• Backend: FastAPI (Python 3.11) + Celery (Workers)
• Frontend: React 19 + Vite + Tailwind CSS 4.0
• Database: PostgreSQL 17 + pgvector
• Cache: Redis 7.4
• Intelligence: Neo4j (Knowledge Graph) + Temporal (Durable Workflows)

---

🧪 Testing

make test          # Run all tests
make test-backend  # Backend only
make test-frontend # Frontend only

---

🤝 Contributing

1. Fork the repo
2. Create your branch (git checkout -b feature/AmazingFeature)
3. Commit changes (git commit -m 'Add some AmazingFeature')
4. Push to branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

---

Built with ❤️ for the NYU community by Harsh Agarwal