3 ms·

Hiring does need to avoid hiring idiots, but I've found that it's more effective to ask easy questions. Seriously. People who can't program a digital computer

Hiring does need to avoid hiring idiots, but I've found that it's more effective to ask easy questions. Seriously. People who can't program a digital computer have just as much trouble, and there's less opportunity to fake it. Easy questions produce better signals.

LordHumungous 11y ago
I think the best questions are those that have an easy initial answer, but which can be expanded upon into a harder and harder direction. This way you are able to measure candidates in a repeatable way based on how far they get.
Klathmon 11y ago
I was asked once during an interview to describe a specific problem that I enjoyed solving. And I was told to go in as much depth as I could during the timeslot. I thought that was a great question. It got me talking about something I cared about, knew very well, and showed my normal thought process and development process outside of an interview. It also gave the interviewer plenty of things to ask about, to "quiz" me on if they wanted, or to just make sure I could actually code. If I'm ever in a position to interview someone one-on-one, that's most likely going to be my main question.
- hunterloftis 11y ago
  I've been asked a similar question which would have been perfect for this author. I had recently published a (terrible, but also terribly fun to write) JavaScript physics engine, which the interviewer saw on GitHub. He began with, "how does it work?" and then continued with probing questions, getting deeper and deeper into the various problems I solved while developing the library, especially focusing on tradeoffs between various choices. By opening the discussion on "my turf," he put me at ease in the conversation, but by driving the interview with very technical questions, he also got the answers he needed about my technical ability.
  - bayonetz 11y ago
    This is a great approach. I do this with candidates I interview. I ask for a sample they are proud of and then I just probe away. Has worked very well.
- bcassedy 11y ago
  There are a number of problems with this approach. It's difficult to standardize so personal biases are more likely to creep in. I've asked this question in interviews a number of times, but it's primarily to gauge communication skills rather than technical skills. The biggest problem with using it as a technical screen is that candidates are really bad at choosing a project to tell you about. I've had candidates choose as their problem building out some responsive dynamic front end and their entire solution boils down to, I picked Angular and Bootstrap.

LordHumungous 11y ago

I think the best questions are those that have an easy initial answer, but which can be expanded upon into a harder and harder direction. This way you are able to measure candidates in a repeatable way based on how far they get.

Klathmon 11y ago

I was asked once during an interview to describe a specific problem that I enjoyed solving. And I was told to go in as much depth as I could during the timeslot. I thought that was a great question. It got me talking about something I cared about, knew very well, and showed my normal thought process and development process outside of an interview. It also gave the interviewer plenty of things to ask about, to "quiz" me on if they wanted, or to just make sure I could actually code. If I'm ever in a position to interview someone one-on-one, that's most likely going to be my main question.

hunterloftis 11y ago
I've been asked a similar question which would have been perfect for this author. I had recently published a (terrible, but also terribly fun to write) JavaScript physics engine, which the interviewer saw on GitHub. He began with, "how does it work?" and then continued with probing questions, getting deeper and deeper into the various problems I solved while developing the library, especially focusing on tradeoffs between various choices. By opening the discussion on "my turf," he put me at ease in the conversation, but by driving the interview with very technical questions, he also got the answers he needed about my technical ability.
- bayonetz 11y ago
  This is a great approach. I do this with candidates I interview. I ask for a sample they are proud of and then I just probe away. Has worked very well.
bcassedy 11y ago
There are a number of problems with this approach. It's difficult to standardize so personal biases are more likely to creep in. I've asked this question in interviews a number of times, but it's primarily to gauge communication skills rather than technical skills. The biggest problem with using it as a technical screen is that candidates are really bad at choosing a project to tell you about. I've had candidates choose as their problem building out some responsive dynamic front end and their entire solution boils down to, I picked Angular and Bootstrap.