Hacker News new | ask | show | jobs
by gdltec 2986 days ago
A developer becomes a "great developer" when the company, team, resources, projects, recognition, etc., are compatible with that person. Under that logic, I believe any programmer can be great if they desire to do so and find the environment and motivation to thrive.

Most technical interviews fail to find the right people because interviewers and hiring managers usually go at it with an "idea" of what a "great developer" looks like to them. In most cases, everyone ends up hiring people who don't work out and miss out on people who could have become the "great developers" there were looking for in the first place.

2 comments

I will second this because I've seen it happen, both to me and to other people -- an environment that wasn't a good fit changes and suddenly someone excels, or someone excels and the environment changes and suddenly it's like they can't do anything right.

Also, having clear criteria set up in advance is essential. You should know before you go into the room (or pick up the phone, or whatever) not only what constitutes a pass versus a fail, but also what kinds of things in the interview signal more than just passing. And the pass/fail can't just be "regurgitated an algorithm we wanted them to regurgitate", because that tells you nothing about whether somebody's actually good to work with.

I agree.

I have a strong belief that the best way to interview a candidate is to take a literal real world problem the team experienced, distill it down to its essence so it can be tackled in a 2-4 hour pairing session. You should have full access to internet, whiteboard, etc.

If the candidate is having to review CS algos and doing whiteboard prep, you're doing it wrong (unless that stuff truly is what the job entails). If the candidate is nervous because they feel like they're giving a dissertation, you're doing it wrong. It should mimic as closely as possible what the actual job will be like.

You're not going to fully know whether or not they will produce in practice, but you'll be able to tell that you can work with them and that's half the battle.