Understanding MLOps -- Initiating the Uninitiated

There is rarely "one pipeline" to manage an E2E process. And there is rarely "one article" to understand a new concept. Understanding MLOps & deciphering jargon such as CI/CD/CT, automation, & deployment rely heavily on the context of our workflow architecture. Deploying a machine learning model into production can involve multiple pipelines that contribute to one large data science workflow. For instance, a Data Engineer prepares the data by sourcing it from a data lake and this is absorbed into the data catalog.