It's going to be fun...
patching servers that are offline with no network connections but have to be powered up and patched monthly due contractual requirements.
collecting audit and other artefacts from systems on a regular basis from 400 segmented servers and appliances
carrying out firmware updates in isolated networks/DMZ's (or on customers who don't have any distribution servers for firmware/driver patching)
building ESX (and other) servers which have no removable media (for security and cost reasons) before they are added to a network
performing disaster recovery on isolated systems because you have to recover their entire environment due to them managing security and anti-virus (badly) in-house