The Courseteer

An Australian undergraduate course aggregator and search engine. Students currently have to visit a dozen different university websites to compare courses, fees, and ATAR requirements. The Courseteer pulls all of that into one place.

How it works

• An AI-adaptive scraper collects course data (name, faculty, campus, ATAR, fees, prerequisites, duration) from Australian university websites
• If a university redesigns their site and selectors break, the scraper uses an LLM to re-map fields automatically
• A fast, filterable search interface lets students compare courses across universities in one view

Tech stack

Layer	Technology
Frontend	Next.js (App Router), Tailwind CSS, next-themes
Database	PostgreSQL (Supabase)
Scraper	Python, Playwright
Scraper AI	Local LLM via Ollama (fallback: Claude Haiku)

Scraper coverage

Phase 1 — Victorian universities (complete)

University	Status
RMIT	Complete
University of Melbourne	Complete
Monash University	Complete
La Trobe University	Complete
Swinburne University of Technology	Complete
Federation University	Complete
Victoria University	Complete
Australian Catholic University (ACU)	Complete
Deakin University	Complete

Phase 2 — All other Australian universities

University	Status
University of Sydney	Planned
University of Queensland	Planned
All remaining universities	Planned

Local development

Prerequisites: dbmate

1. Environment setup:

   cp .env.example .env
   # Set DATABASE_URL to your Supabase connection string (sslmode=require)

2. Database: This project uses Supabase for both local dev and production. Point DATABASE_URL at your Supabase instance — Transaction Pooler on port 6543 is recommended.

3. Run migrations:

   scripts\migrate.bat        # Windows
   bash scripts/migrate.sh    # macOS/Linux

   Migrations live in db/migrations/. dbmate tracks applied migrations and reads DATABASE_URL from .env.

4. Run the scraper:

   cd scraper
   uv run python run.py --university rmit

Project status

Phase 1 (Victorian universities) is complete. Phase 2 expands coverage to all remaining Australian universities.

Component	Status
Database schema	Complete
Scraper infrastructure	Complete — robots.txt compliance, snapshot caching, AI re-mapping
Victorian university scrapers	Complete — all 9 universities
Next.js frontend	Complete — course listings, filters, pagination, campus/ATAR display
Admin health dashboard	Complete
Authentication (multi-provider)	Complete
Phase 2 scrapers	Planned

License

MIT — see LICENSE.