Visual AI QC - Repository Structure

⚠️ Important: Work in backend/ Directory

All development and production code is now in the backend/ directory.

Directory Structure

ai_qc/
├── backend/              # 👈 MAIN APPLICATION - Work here!
│   ├── api_server.py     # Flask application
│   ├── llm_config.py     # LLM configuration (Gemini model settings)
│   ├── profile_config.py # Profile management
│   ├── visual_qc_apps/   # QC check modules (48 checks)
│   ├── profiles/         # QC profile configurations (9 profiles)
│   ├── brand_guidelines/ # Reference assets
│   ├── scripts/          # Deployment scripts
│   ├── config/           # Environment configs
│   ├── uploads/          # File uploads (production)
│   ├── output/           # Generated reports (production)
│   └── ...               # All other application files
│
├── frontend/             # Frontend-only deployment (optional)
├── config/               # Legacy config (deprecated)
└── CLAUDE.md             # Development guide

Quick Start

Local Development

cd backend/
./scripts/run-local.sh
# Access at http://localhost:7183

Production Deployment

# SSH to production server
ssh user@server

# Deploy backend to /opt/ai_qc
cd /opt/ai_qc
git pull origin main
cd backend
sudo systemctl restart ai_qc.service

Why This Structure?

Previously, we had duplicate files in both root and backend/ directories, causing sync issues. Now:

✅ Single source of truth: All code in backend/ ✅ No sync issues: No duplicate files to keep in sync ✅ Clear structure: Production deploys from backend/ ✅ Simpler workflow: One place to make changes

Documentation

• Full Documentation: See backend/CLAUDE.md
• Deployment Guide: See backend/DEV_PROD_SETUP.md
• API Endpoints: See backend/README.md

Making Changes

1. Navigate to backend/
2. Make your changes
3. Test locally with ./scripts/run-local.sh
4. Commit and push
5. Deploy to production: git pull on server

---

🚀 Always work in the backend/ directory!