Answers

1. We switch the localhost part of the URI to the tag of the docker-compose service.
2. Configuration files enable us to switch contexts easily; for instance, switching from development to live. Additionally, if our Celery service needs to talk to a different database for some reason, this can be done with minimal effort; simply changing the configuration file will work. It is also a security issue. Hardcoding database URIs will expose these credentials to anyone who has access to the code and will be in the GitHub repository history. Store the configuration file in a different space such as AWS S3, which gets pulled when the service is deployed.
3. Technically, no. We can simply write SQL scripts and run them in sequence. When I was working in financial technology, this was actually a thing that we had to do. While this can give you more freedom, it does take more time and is more error-prone. Using alembic will save you time, errors, and work for pretty much most of...