2025 Meta Data Engineer Onsite

andy3545 发表于 2025-02-19 06:33:03
If I remeber correctly, it's something like this

ratings =

Is the rating a list or a dictionary? Will there be duplicated keys in it like this?

Thanks for sharing!

Hihi, very helpful and detailed post, thanks so much
Can you tell us a little more about Full Stack 3 (Indian Interviewer) regarding the python stream data processing problem, what does the input look like and what do you need to return as a result? what are some edge cases?

andy3545 发表于 2025-2-19 09:33
If I remeber correctly, it's something like this

ratings = {movie_1: 3.5, movie_2: 4,  movie_3: ...

if ratings is a dictionary, the key in dictionary can not be duplicate. the value will be overwritten by the last one.
should ratings be a list?

For the sql part, does meta use presto version for interviewing?

MalikJia 发表于 2025-3-5 17:55
For the sql part, does meta use presto version for interviewing?

It doesn't matter since you wont be able to execute any code. But Meta uses ANSI sql (presto/trino)

zombie8 发表于 2025-3-3 21:51
if ratings is a dictionary, the key in dictionary can not be duplicate. the value will be overwrit ...

I think it was dictonary with value as lists to store multiple ratings.

Were there team match when you get the offer?

匿名用户 发表于 2025-3-10 01:47
Were there team match when you get the offer?

Since I was referred by a friend, I skipped team matching and joined their team directly. I also had the options to talk to 3 other teams.

thanks for sharing ! added rice
My HR told me that for three product interview,  there will be two real meta products and one makeup product. It looks like the ride-share one will be the makeup one. Does that means the video streaming platform should be considered be lunched on meta as well?

Thanks for sharing, for Ride-sharing design, are we supposed to design a data model for OLTP or OLAP