Missing out of jobs because of LeetCode

At the end of the last year, my previous workplace ( Weaveworks ) went bankrupt. I found myself without a job. I didn’t have to interview for a long time and I certainly wasn’t planning on it any time soon. I enjoyed my work.

Thus, I started to learn a bit because I forgot how to interview. I refreshed my knowledge on LeetCode, I did some exercises, I read some books, post, whatever. I refreshed my algo knowledge because that’s what people most likely will ask, right?

Well, no. I had several interviews over the course of months I was looking. Some where okay, some were not so great. Some, where outright horrible.

Nevertheless I realized that I neglected to refresh my knowledge about these key areas:

  • distributed systems
    • mostly about locks, handling channels in go, doing some exercises like the dining philosophers and the likes
  • kubernetes internals
    • this one was interesting; I worked in this area for such a long time that I forgot details that I took for granted
    • like, I was doing some things with controllers but I forgot why…
  • architecture and design of more complex systems
    • I read the primer for this, but I neglected the details and got too lazy over the years. I seldom had to do that (designing something from scratch) so I was missing some key points

But most importantly, apparently, I appear as nervous and agitated. Well. Color me surprised. I had no job, I ran on my savings, and the workplace I was interviewing for looked nice. OF COURSE I WAS NERVOUS.

In any case, I focused so much on algorithms and coding, that I neglected the important bits and pieces of my knowledge. The pieces that actually matter. And the pieces good workplaces actually care about.

I’m not saying this is a perfect score. There are workplaces that will quiz you on LeetCode or some other meaningless algorithm. And telling them that, yes I know about it, but I would rather just read a book than try to figure it out and implement it from scratch; or better yet, I would use a library that has peak implementation of said algorithm - simply just never works.

Which is sad. Because it should. Workplaces should look for people who understand the relevant technology or show promise in learning about it or worked with something related.

Algorithms are not the future unless you work as a trading savant or you specifically work in algorithms research.