|
|
|
|
|
by vidarh
1419 days ago
|
|
Turned down Google receuiters way too early to know if that's typical, but Facebook recruiters told me from the outset the process was 6 months and that they'd provide me with a reading list, and was then surprised when I told then I'd let them know if I was willing to consider that if they provided me with a compensation range. Days after, once they'd figured out the comp range, I told them I wasn't interested. This is London, and while their total comp was good, it was not SV good and not a big enough step up to be worth 6 months of interviews and a reading list. |
|
That's crazy. When I joined FB a few years ago the total time was under a month.
The reading list thing makes sense for some people who have never done a FAANG-like interview before. I remember moving from London to the USA early in my career and I had no idea how the process worked and so I got my first phone screeen with Amazon, thinking I'd explain a bit about my background and we'd have a bit of a chat and instead I got "Okay. Give me code to perform an in-order traversal of a binary tree." which was NOT something I was remotely prepared to do at the time :)
If you know the format and basic types of questions (and have access to leetcode etc.) it's not too bad, but a lot of these resources weren't around 15 years ago, especially to people from another country with no context.
A second Amazon interview aside: in an in-person interview I did, things had been going very well and I was on my fifth interview of the day. Again, I was recently from London and hadn't fully understood that words mean different things in the USA than they do in the UK. I was asked to design a GPS system. This seems simple enough if you know that "GPS" in the USA is the same thing as a "SatNav" or "Satellite Navigation" system in the UK and that the goal would be to build a system that could compute an efficient route between two points.
The problem was that I assumed the person wanted, well, a GPS system. i.e. he wanted a design for a set of satellites in geosynchronous orbit that would, using very precise timing, allow a person almost anywhere on earth to get a latitude and longitude like "12.345, -2.345" and so we spent almost the whole of the conversation talking about two different things and even when I was given the hint that maybe I could use some kind of graph algorithm, I had no idea how to use Dijkstra's to solve clock drift on a satellite :)