AI Workflow Builder (concise)

This repository contains a prototype full-stack AI Workflow Builder with:

• FastAPI backend (camera ingestion, ONNX detector adapter, DB models)
• Vite + React frontend (Camera Dashboard)

See workflow_documentation.md for the complete design and architecture notes.

Quick start (local)

Backend

1. Create & activate a Python virtualenv:

cd backend
python -m venv venv
source venv/bin/activate

2. Install Python dependencies (example):

pip install -r requirements.txt
pip install "uvicorn[standard]"

3. Run the backend:

uvicorn main:app --reload --host 0.0.0.0 --port 8000

Frontend

cd frontend
npm install
npm run dev

Notes

• ONNX and OpenCV are optional: install onnxruntime / onnxruntime-gpu and opencv-python-headless if you want real inference instead of the mock mode.
• Remote model downloads are cached in backend/.model_cache and that folder is ignored by git.

License: MIT