Sofar it’s seen that even if a new parent industry doesn’t drum up as many jobs as the one it’s replacing, the supporting industries fill the gaps, if not exceeding the original amount. Take airlines for example, not that many people comparatively are pilots, but many people work in tourism, as stewards/stewardess, as support staff at airports, etc.
I would imagine if we could perfect automation to the point that it could assemble anything better than humans, we would indeed all be focused on service industries, or living off of some form of basic income system.
Automating many service sector jobs is hard, and some impossible, as humans need face to face interaction, and generally do not enjoy the rigidity of talking to a machine (How many times have you called a company only to think “please just give me a real human”).
I’m in the civil engineering field myself. I don’t fear automation at all. Sure, you could automate a lot of structural design, and it essentially is already thanks to analysis software, but at the end of the day you need someone who knows what they are doing to make sure the numbers look right and nothing seems out of place.
Re self checkout, that also generates many more jobs then it destroys. Now you’ve got programmers, designers, engineers, tech support, service staff, that all must exist to maintain this industry.