The mlops directory contains all the logic and configuration to operationalize your data science project.
The python scripts in is folder contain all the code that manages the end-to-end orchestration of ML artifacts through Azure Machine Learning. For instance, sending the training to a remote target, creating data science pipelines, using automl, etc. Basically, everything functionalities that is not core data science scripts.
The aml_utils folder is a helper module used by these mlops scripts to interact with Azure Machine Learning through the SDK.
The tests folder contains different tests to be run in the CI/CD pipeline: unit tests, integration tests, etc.