The Lounge is rated PG. If you're about to post something you wouldn't want your
kid sister to read then don't post it. No flame wars, no abusive conduct, no programming
questions and please don't post ads.
I am conducting a round of interviews this week, looking for a mid-level C# developer. I'm down-selecting some tech questions which are aimed at trying to filter out the "Yeah, I did a C# project a couple years ago" crowd from the "C# has been in my blood for at least a year or two" folks.
But selecting appropriate questions is always a challenge. (Just because something is obvious to me, doesn't mean every developer has encountered it. Every project is different and people get exposed to different things. I definitely want to avoid "language trivia.") So here is my first cut of questions. For those of you who are .NET devs, would you agree that somebody with a couple years of solid experience should be able to talk intelligently about most of these topics? (Not all of these are C#-specific.)
How do you inherit a class, and why might you want to?
What is the purpose of interfaces, and how are they helpful?
Why would you ever want to make a method private or protected?
When might you use a static class or method?
Can you explain what a lambda is, and why you might use one. (Or, alternatively, LINQ?)
Why might you use a property instead of a regular variable? (How are they different?)
Are you familiar with any "Design Patterns"? Can you name one or two that you have used?
Have you heard of the concept of "tight" or "loose coupling", and how does it effect code design?
Have you used a Unit Testing framework? If so, how did it (or unit testing in general) benefit your code, if at all?
[Note: we have decided not to make the candidates write or debug actual code in the interview, with the possible exception of FizzBuzz. But that is a topic for another thread.]
Really? I particularly liked those questions. What I liked about them was that, for the most part, there isn't really any right or wrong answers, per se, but I would be surprised if somebody had been doing development work for a year or two and hadn't encountered them in some form. I don't care about the exact syntax; I'm just interested to hear if they can discuss the concepts thoughtfully.
Not a bad set of questions at all - nothing too narrow in there.
If not getting them to write code, it might be a good idea to ask how they would approach writing something for a stated problem. In our last round of interviews (admittedly for a junior), I set what I considered to be a super-easy coding task. I thought at the time that it would filter out a couple of idiots but hardly any candidates even got close to it.
The task was simply to take an input of a sum of money and break it into the smallest number of notes and coins possible.
The first attempt was hilarious (pretty much if (amount == 52.31) print "some answer that added up to about £48") and they didn't get much better from there.
At one point I was so worried that I'd gone OTT that I asked my ageing mother. She worked out the algorithm in half a second.
The thing was, some of these guys could talk the talk (or at least quote Wikipedia) but couldn't code their way out of a wet paper-bag so I really would go with a test of that sort.
Seriously, though. I think that's kind of the problem with interviews: bright people can get so nervous that they aren't able to think as quickly or clearly as they might under normal working conditions. Good/bad interview performances don't always correlate to actual skill or aptitude. That makes it tough for an interviewer.
These are pretty good questions for a mid-level developer.
What I have found to be the best way to gauge a developer's skills is to ask them to tell you about different projects they have worked on. If they don't give much detail ask if they worked on it alone, if they wrote all the code or just maintained it, if they designed the classes etc. You should be able to tell if they are qualified by allowing them to do most of the talking.
There are two kinds of people in the world: those who can extrapolate from incomplete data.
There are only 10 types of people in the world, those who understand binary and those who don't.
I can completely answer 8 out of 9 having used C# poorly in a single project 5 years ago. An that's because I don't remember Design Patterns, not by name - I always used a bunch of them even before being introduced to the concept of design patterns... When I read that infamous book I mostly thought "well, duh!" every odd page.
You'll get someone good at exams, doesn't necessarily translate to ability to create solutions from scratch.
The Asian's education departments are fighting this very weakness in their current system right now, they got plenty of kids who ace exams but are useless in the field without constant mentoring. (Those 14 year old geniuses doing PhD's - touted to be the next Einstein and will solve world peace, hunger and global warming - never heard from again.)
Ask them not "how do you ...," rather questions where they will need to suggest use of such items as you listed. If answers are vague ask for details. i.e. simple case: don't ask a builder "how to use a tape measure," ask them how they would determine the width of a doorway.
Sure there are other ways to do things, if it sounds odd ask them why they chose that way, challenge them on the got-ya's, ask why they believe their solution is the best way and what's the alternatives. (Warning: Don't play teacher and correct them - that's just a recipe to become their mentor if hired - OK to throw in the odd clue but not too much. Act like a doctor, throw in a few hmmm's and take notes - that last one will get them sweating and acts as a good stress test.
Sin tack ear lol Pressing the any key may be continuate
Last Visit: 31-Dec-99 19:00 Last Update: 29-Mar-17 7:36