One solution to all of this is for the applicant to do a few things to avoid being confused for a terrible engineer. The last time I looked for a job, more and more companies were asking for an engineering design portfolio. Instead of waiting for the employer to ask you to talk intelligently about things you haven't learned yet, show up to the interview with a portfolio of things you can discuss on a detailed level for hours on end.
I've only been working professionally for a little under 4 years, but I've encountered a handful of terrible engineers - engineers who only know one solution and apply it to every problem that presents itself, very smart, technical people who know enough to hack something together and leave the difficult, last 20% for other people to solve, etc.
Another issue, especially in larger companies, is you have to allow a bad engineer to fail in order to justify firing them. In a fast paced environment, sometimes you can't allow a project to fail simply to provide cause to fire an engineer. And even after the failure, HR will require you to implement a performance improvement plan, etc.
The problem with the portfolio approach is that many incompetent engineers can easily point to code that they "wrote." Often it was part of a larger team carrying them, but sometimes they wholesale copy it.
A similar problem applies to pure project interviews: the total incompetents are also the ones most likely to get someone else to "help" them with it.
Yep, there also needs to be a way to detect bat-boy syndrome (where the team goes to the world series, but the guy you interviewed was just there to lug the equipment around).
I've always heard that you learn more in an hour of playing(as in a sport) with someone than you will in ten hours of conversation. I like the idea of being paired with an engineer to solve a problem because it let's someone see how you think, what your process is, and, most importantly, how you learn.
>The problem with the portfolio approach is that many incompetent engineers can easily point to code that they "wrote." Often it was part of a larger team carrying them, but sometimes they wholesale copy it.
Have them walk you through it, or ask specific questions about it.
>A similar problem applies to pure project interviews: the total incompetents are also the ones most likely to get someone else to "help" them with it.
The best interview I've ever seen is where they pair you with another engineer and give you an hour or 2 to solve a problem. The other engineer is there to help you and you actually work as a team.
I've only been working professionally for a little under 4 years, but I've encountered a handful of terrible engineers - engineers who only know one solution and apply it to every problem that presents itself, very smart, technical people who know enough to hack something together and leave the difficult, last 20% for other people to solve, etc.
Another issue, especially in larger companies, is you have to allow a bad engineer to fail in order to justify firing them. In a fast paced environment, sometimes you can't allow a project to fail simply to provide cause to fire an engineer. And even after the failure, HR will require you to implement a performance improvement plan, etc.