This project contains simple starter code I've used on various projects, including an Autograder for IU's Intro to AI course I created.
This project is a sample React.js app that is served from a Flask app. It uses GitHub OAuth for authentication and SQLAlchemy as an ORM. The GitHub OAuth service can be changed to virtually any with ease.
The front-end is a CRA React.js app, which you can find in the client_app directory. The back-end, found in the server directory, is built in Python and uses Flask w/ SQLAlchemy.
In order to run locally, first start the React app by running npm start while in the client_app directory. Then, start the flask app by running run.py. Don't forget to configure settings found in server/settings.py. You will need to create a .env that contains the environment you're running the app under. See the example included for reference. The Flask app contains several comments on how it all works. I recommend starting by looking at server/__init__.py.
The flow I use in order to publish the app on a development or production environment looks like the following: build the react-app by running npm run build. This puts the built files in client_app/build. The site blueprint references that directory in order to server the app to the client. Configure the environment set in the .env (i.e., PROD). Start the flask app however you'd like, I usually prefer gunicorn. Finally, I use a NGINX reverse-proxy that points my domain (i.e., https://example.com) to the gunicorn server (i.e., localhost:5000).
You can do this however you'd like, refer to Flask's deployment options documentation for more info.
I currently don't have this setup anywhere, but will add a link once I do.