Is There a Machine Learning Pipeline?

The way I see a pipeline, especially in the data context, is really a way how you can construct a sequence. And that sequence can also have parallel steps of jobs, right? It's a day after you're constructing. The concept of the deck is the same. You're still composing your deck. But what you get out of the box probably helps you do things a lot faster than you would do the same thing compared to you doing the same thing with airflow.