Sanyamgoyal21/Modelflow ? reverse-engineered prompt

Build me a web app called Modelflow that lets a user log in with Google, upload a machine learning model, and get back a live API endpoint without having to set up servers or Docker themselves.

I want a simple dashboard where people can see their uploaded models, deployment status, and the API details for each one. The app should validate the upload, package the model into a container, and prepare it for deployment automatically. It should also show basic info like whether the model is ready, running, or failed, and make the public endpoint easy to copy and use.

Use a modern full stack setup with a React frontend, a Node and Express backend, MongoDB for storing user and model data, and a small Python inference service for the model runtime. Include the pieces needed for secure auth, file upload, and deployment flow, and wire it up so it feels like a real product. If you need current docs for any cloud or auth setup, look them up online first.