Sphinx-SCA

IntelliMath AI (Sphinx-SCA)

Multi-page Vite frontend + FastAPI backend for an AI-powered math solver.

Quick start (local development)

1) Configure environment variables

• Copy .env.example to .env
• Fill in at minimum:
  • Frontend: VITE_SUPABASE_URL, VITE_SUPABASE_ANON_KEY
  • Backend: GROQ_API_KEY
  • Optional: SCOUT_API_KEY (enables image understanding)

2) Run backend (FastAPI)

pip install -r requirements.txt
uvicorn backend.app:app --reload --host 0.0.0.0 --port 8000

Backend health check: GET /health

3) Run frontend (Vite)

npm install
npm run dev

The frontend uses the dev proxy in vite.config.js for:

• /solve_stream
• /ocr
• /api

So in local dev you can leave VITE_API_URL empty.

Production build

npm run build
npm run preview

Deployment

Render (recommended, already configured)

The repository includes render.yaml defining:

• math-backend: Python web service
• math-frontend: Static site built from Vite dist/

Set the required env vars in Render:

• Backend: GROQ_API_KEY (required), SCOUT_API_KEY (optional), ALLOWED_ORIGINS (recommended)
• Frontend: VITE_SUPABASE_URL, VITE_SUPABASE_ANON_KEY

VITE_API_URL is automatically wired in render.yaml from the backend external URL.

Notes

• Do not commit .env files. Use .env.example as the template.
• dist/ is build output and should not be committed.

Developer docs

• Mathematical output + agent best practices: docs/AGENT_AND_OUTPUT_GUIDE.md