I haven't had a job interview without leetcode bullshit in about 15 years. And years before that.
Basically since Joel's posts in the early 2000s. He promoted a reasonable idea-that programmers should know how to program. Then B and C players rushed to make that happen!
Similar to how agile became the complete opposite of what it meant originally.
I was an interviewer at [FAANG]. I did 300+ interviews. At the beginning I asked tough questions since I didn't know better, but I quickly realized I was not getting the signal I wanted. So I changed and just asked simple questions that included writing some code. There was no gotcha, no need to know sophisticated data structures. Leetcode practice would not have helped much.
You may be focusing on the leetcode label, and not the rest of the fail-fast zingers that are ubiquitous. Only a subset are actual leetcode.
The first mistake and you're done, due to the high number of applicants.
If you've got networking to let you bypass the BS, that's great. But most interviews (by number) are "cold" interviews that conduct these almost exclusively. Due to lack of trust and extreme fear of failure.
The only offer I've received in years was because, while they did this type of test as well, they didn't take the results too seriously. One out of maybe a hundred attempts that year.
Basically since Joel's posts in the early 2000s. He promoted a reasonable idea-that programmers should know how to program. Then B and C players rushed to make that happen! Similar to how agile became the complete opposite of what it meant originally.