Role: Site Reliability Engineer
I had found a really useful interview report on the forum before mine, and the overlap with what I actually got asked was pretty high. This is also my own area of work, so going into the interview I felt like I had a 100% chance of passing, but I still failed. I'm not really sure why. Let me write up roughly what was asked — honestly you can just ask AI what kinds of questions it can generate around these topics.
Linux: the difference between du/df and why disk usage shows differently between them, signals, kill, zombie processes, the init process, etc.
Python: generators (yield), Python decorators, context managers
Troubleshooting: what to do if you can't SSH into a host, combined with tools like traceroute, ping, ss, lsof, sshd, and authentication
My mindset was wrecked. What more can I say, it's just tears. I'm nowhere near the top performers here, but I answered every single question, and I even discussed a lot of the concepts like signals and tools like traceroute at the level of underlying principles. I used to be the one interviewing a lot of people myself, as an interviewer, and I never expected to crash and burn on the exact thing I'm best at... I guess I still have more to learn.
Discussion
Loading comments…