I was over-simplifying to communicate the repetitive and dull nature of a job I'm very happy to no longer be doing. Really the game was more about "anomaly detection" and the sensor indicators/measurements were far more continuous than categorical, and their outputs had to be weighed against past experience and context of the monitored components.

Well, to start with, you have to use ML to train the computer vision system to recognize green and red lights. After that you update your resume with the ML experience you gained. :)

I want to ask a serious question about the ML, but your answer is great! My experience of anomaly detection, even with deep networks, is that recognition of the important anomalies can be done for some of them which means that you can raise a big flag for those - but someone still needs to sift through everything else.