Hi Paul,
How does using those directions compare to this tutorial?
https://docs.microsoft.com/en-us/azure/machine-learning/data-science-virtual-machine/dsvm-common-identity
Would my understanding be correct that the github directions are focussed on providing AD integration for one single DSVM, while the common identity does this for any DSVM in the subnet, thus providing more flexibility?
Both at the end say that logging into JupyterHub using AD credentials should work.