I've read couple automated themselves out of a job 1 2 3 posts, and I wonder to myself how this is done.
I can understand automating something to save 99% of the time, but I can't imagine setting up a system that can self-regulate and self-correct for future changes. If I had to process data, what happens if I need to process an extra column? What if the computer died? What if the user dragged a .LNK shortcut into the program? etc.
I guess I come from a programming background, so I naturally go through all the failure scenarios and consider whether the system can handle it.
I have done huge amounts of automation before, and the output is very significant (1 untrained person monitoring a system that works at the rate of 10 trained staff), but I still need that one person sitting there to either reboot the computer due to memory leaks, or the internet is slower than usual, or just stop it and someone else runs a query to confirm the results for the last 4 hours of processing. For me to code that final 1% will take an astronomical amount of time, so I usually stop and let a human take over.
Do people preemptively think they are automating themselves out of a job? Or do these "automate yourself out of a job" scenarios really exist?