Our interview process is painfully slow and broken
old.reddit.com
old.reddit.com
For an AIX project on PS/2, they sent me a double stack of AIX manuals. It was an X Windows project so I bought all the O'Reilly X books and read the main two.
The next project was an RS/6000 project. Same code base (Ford's in-house car design program) but now targeted for PHIGS. I had never even heard of PHIGS. Ford already had the code written, but the trickier things like surface shading were not completely standardized, so every vendor had a slightly different way of setting up the data structures. IBM's Graphigs had to be used.
Next project was the same Ford code, but now on Silicon Graphics with a PHIGS library licensed from a French company.
A friend recommended me once for a project where PCs were connected to an IBM mainframe using 3270 emulation software. The company (Brown-Forman) wanted to automate some aspects of their terminal sessions. I used Turbo Pascal to communicate with the PC emulation software. I'd never used a mainframe, mainframe terminal, 3270 emulation software, or Turbo Pascal, though I did know Pascal.
I would have never been hired for any of these projects if I had to know them inside out to pass an interview before even starting.
No wonder there is a shortage of tech talent. Tech companies today expect you to know everything they might want you to work on when you walk in the door. Ever heard of learning?
I'd say 'good luck' but you really need to refactor the recruiting process here