4 ms·

Yeah he better be able to invert a binary tree, else how do we know he can code? /s

eastbound 3y ago
You can’t imagine what we see on a job. People with 20 years experience in programming, perfectly able to talk about it, and then they fail at writing a String.indexOf() manually (requires 2 for and an if).
- CoastalCoder 3y ago
  I'm that guy. After getting laid off, I did legitimately fail some basic programming tests. Some combo of nerves and rustiness, I expect. I've found that practicing leetcode-like problems was necessary and sufficient for fixing that.
  - withinboredom 3y ago
    Same! I remember failing a coding interview because I forgot how to write a for-loop despite using them almost every day. It’s crazy how nerves affect you.
    - johntiger1 3y ago
      Yeah I've failed 'reverse a string' before, despite graduating with close to a 4.00 GPA and getting a 99 in my DS/A course from a Canadian school (Waterloo/UofT). Especially early on in your career, the transition from solving algorithms questions on an exam to leetcode-style is quite a jump, especially under time pressure
- guitarbill 3y ago
  It's tough. I've also seen what you describe, and it's kind of mind-boggling. The only red-ish flag seems to be overly long tenure at a single place. But at some point in the career progression, it doesn't matter. I think the difference is this. Some people I've worked with were excellent coders but have now moved on to e.g. director roles. If you asked them to write code, they'll flat out say that they haven't coded in years. They know it isn't important for their role, and it isn't the value they provide. The bullshitters will never admit they can't do something.
  - johntiger1 3y ago
    Would long tenure at e.g. Meta or Google also be seen as a red flag? I've heard other people say that the opposite (job hopping every 1-2 years) is a red flag.
    - 59nadir 3y ago
      The average tenure at Google is apparently 1.1[0] years, so it's at the very least not typical to stay for very long. 0 - https://stackoverflow.blog/2022/04/19/whats-the-average-tenure-of-an-engineer-at-a-big-tech-company-ep-434/ https://stackoverflow.blog/2022/04/19/whats-the-average-tenu...
      - aitchnyu 3y ago
        Umm, do these companies factor average tenure to tweak vesting schedule and finetune how much stock they fork over to employees?
      - jsnell 3y ago
        Tenure isn't a good way to measure how long people stay. Median tenure (which this is reported to be) is even worse. All a median tenure of 1.1 years days is that about half the workforce was hired in the last 13 months, but it could be either due to the number of employees growing rapidly, due to people leaving the company, or some mix. We can't tell which just from these numbers. As a thought experiment, consider a company that nobody ever leaves, growing by 20% per year. They'd double in size in about 4 years, so the median tenure in steady state is also 4 years. But that doesn't mean that people only stay for 4 years; the turnover in the thought experiment is literally 0%. Also, the numbers the article (podcast?) is claiming are not supported even by their own immediate source. The reference they have for average tenure for software developers being 4.2 years doesn't actually say that. The article actually says that it's 4.2 years for all employees and that sofware developers are about 2 years. It also has such a different number for Google (2.3 years) than the number they chose to originally quote that the methodology just can't be the same. I would guess that the second source is using mean, not median.
- guappa 3y ago
  Having experience isn't an indicator of knowing how to do basic things, because companies hire just about anyone.
- johntiger1 3y ago
  Were you asking them in a language they were familiar with? They also might have gotten nerves and thought about using some fancy DP algorithm and gotten stuck in that rabbithole. I've personally definitely had my fair share of unbelievable brain farts

You can’t imagine what we see on a job. People with 20 years experience in programming, perfectly able to talk about it, and then they fail at writing a String.indexOf() manually (requires 2 for and an if).

CoastalCoder 3y ago
I'm that guy. After getting laid off, I did legitimately fail some basic programming tests. Some combo of nerves and rustiness, I expect. I've found that practicing leetcode-like problems was necessary and sufficient for fixing that.
- withinboredom 3y ago
  Same! I remember failing a coding interview because I forgot how to write a for-loop despite using them almost every day. It’s crazy how nerves affect you.
  - johntiger1 3y ago
    Yeah I've failed 'reverse a string' before, despite graduating with close to a 4.00 GPA and getting a 99 in my DS/A course from a Canadian school (Waterloo/UofT). Especially early on in your career, the transition from solving algorithms questions on an exam to leetcode-style is quite a jump, especially under time pressure
guitarbill 3y ago
It's tough. I've also seen what you describe, and it's kind of mind-boggling. The only red-ish flag seems to be overly long tenure at a single place. But at some point in the career progression, it doesn't matter. I think the difference is this. Some people I've worked with were excellent coders but have now moved on to e.g. director roles. If you asked them to write code, they'll flat out say that they haven't coded in years. They know it isn't important for their role, and it isn't the value they provide. The bullshitters will never admit they can't do something.
- johntiger1 3y ago
  Would long tenure at e.g. Meta or Google also be seen as a red flag? I've heard other people say that the opposite (job hopping every 1-2 years) is a red flag.
  - 59nadir 3y ago
    The average tenure at Google is apparently 1.1[0] years, so it's at the very least not typical to stay for very long. 0 - https://stackoverflow.blog/2022/04/19/whats-the-average-tenure-of-an-engineer-at-a-big-tech-company-ep-434/ https://stackoverflow.blog/2022/04/19/whats-the-average-tenu...
    - aitchnyu 3y ago
      Umm, do these companies factor average tenure to tweak vesting schedule and finetune how much stock they fork over to employees?
    - jsnell 3y ago
      Tenure isn't a good way to measure how long people stay. Median tenure (which this is reported to be) is even worse. All a median tenure of 1.1 years days is that about half the workforce was hired in the last 13 months, but it could be either due to the number of employees growing rapidly, due to people leaving the company, or some mix. We can't tell which just from these numbers. As a thought experiment, consider a company that nobody ever leaves, growing by 20% per year. They'd double in size in about 4 years, so the median tenure in steady state is also 4 years. But that doesn't mean that people only stay for 4 years; the turnover in the thought experiment is literally 0%. Also, the numbers the article (podcast?) is claiming are not supported even by their own immediate source. The reference they have for average tenure for software developers being 4.2 years doesn't actually say that. The article actually says that it's 4.2 years for all employees and that sofware developers are about 2 years. It also has such a different number for Google (2.3 years) than the number they chose to originally quote that the methodology just can't be the same. I would guess that the second source is using mean, not median.
guappa 3y ago
Having experience isn't an indicator of knowing how to do basic things, because companies hire just about anyone.
johntiger1 3y ago
Were you asking them in a language they were familiar with? They also might have gotten nerves and thought about using some fancy DP algorithm and gotten stuck in that rabbithole. I've personally definitely had my fair share of unbelievable brain farts