Latest work = JsonFixer

AI is the reason interviews are harder now

15 May 2024

For past few weeks, I've been prepping for interviews. One thing that caught my eye was the difficulty of questions have dramatically increased.

In 2021, you were expected to solve a question in DP/recursion/trees etc in 45 minutes. which is tough by itself.

But in 2024, companies are giving leetcode hards and asking to solve them in less than 30 minutes.

Example 1 [asked by a FAANG]:

parallel-coursesThis question requires DP on graphs with bit masking and backtracking. (Maybe topological sorting too.)

Example 2 [asked by Uber] :

kenken-puzzleHere's what KenKen puzzle wiki looks like. Good luck doing this in half an hour!!

Who solves these Questions?

Well ofcourse. It's the people using:

and other similar tools.


The creator of Ultracode said it best:

Ultra code is well equipped to answer any techical question you throw at it. Algorithmic, OOP, System Design. It doesn't matter. Even someone non-technical like myself can pass a sweet interview.
38:20 in https://www.youtube.com/watch?v=qJQQ-0BQNOc

People don't actually use these right?

See for yourself:

blind-1blind-2

What does this mean for actual studying peasants

We would be continually outcompeted unless we start using these tools. A true tragedy of commons.interviewing.ioSource: Interviewing.io

People using GPT are regularly caught

The idea of spotting cheating is obvious is a case of selection bias. You only notice when it's obvious. Clearly, the person put 0 effort towards cheating (as most cheaters would, to be fair). But using the above tools, or even just having some sort of setup would make the issue much harder to spot.

Solution

Do the interviews in person. There's no other way.

Should Interviewers care

YES. Because AI can't do all functions of a software developer (yet). This means the phony guy you hire will have to do them. and you have to oversee the activity.



Thanks for Reading!!