Improve ability of end users/AIs to develop Babel components

[gaurav]

But how? Some options:

• Write a Jupyter Notebook template for creating an ingest that can be completed for a particular data source and then e-mailed to us to test and incorporate.
• Set up some kind of private Jenkins instance where a particular job on a particular branch of Babel can be run remotely.
• Solicit in-house folks to try running Babel locally and to document pain points that we can fix.

This is in addition to the other tooling we're building to work with, such as #562