After my undergrad freshman year: Working for Atari repairing 8080 uP boards for the Space Invaders game console (the big ones). They wanted to save money buy buying cheaper PCBs from somewhere. They were crap, open traces, shorted traces, etc. This was well before anything like a bed of nails test fixture. So they just built them up and the ones that didn't work were sent to us.
My tools were a "game console" test aid (that looked nothing like the real thing, just a bank of switches and such), a DVM, and a basic oscilloscope. The best (normal) troubleshooting method involved cutting traces or IC pins, measuring things, and repairing them with solder bridges. Something you'd get instantly fired for anywhere that cared about reliability. But they'd ship anything that would work for an hour.
But the best, and, after a while, the worst part was that you had to play the game to fully test it. So after a day or so I was elated that they were paying me to play a video game. The problem was, it was the same game over and over again. It's a fairly simple game to master, and as long as you don't fall asleep, you can't ever lose. I learned to hate Space Invaders. The real game was the troubleshooting, not the verification testing.
It was an easy way to earn money though. I also learned a lot about how to work with simple uP systems since everything (RAM, ROM, PICs, etc.) was separate ICs connected by potentially faulty traces. Maybe like a doctor feeling your pulse; I could move my scope probe down the address and data buss and often get a quick idea of the nature of the problem(s). I'd guess it's an unnecessary, lost, art now. Those were primitive days in the uP world.