To start the project, start a Redis docker instance using this command:

docker run -d --name redis \
  -p 6379:6379 \
  redis:7-alpine

Then start the four sub-folders frontend, primary-backend, worker and ws, following the README file in each one to start it up.

My recommendation is to add to the database in worker, as a component of the processing function. It allows for the maximum amount of progress before failing out, if something goes wrong.

As for how to deploy, I don't really know. My inclination is to say as a monorepo, but that is only how I would do it, not strictly the most effective way.