Hard Interview Problems

1. Apr 21, 2013

Monte_Carlo

Over the years I've heard gruesome stories about engineering/science/tech interviews - questions that they expect you to get on the spot (or nearly so) but which would require me to spend a good day thinking. I've already read Poundstone's books but I have to say there are very few good ones like that out there about interviews/relevant puzzles (any suggestions are welcome.)

While I know it may have to do with specific knowledge (like propulsion) I suspect even in aerospace they sometimes test applicants with generic diffucult analytical problems. I would appreciate visitors of this forum to post some difficult STEM interview questions (i.e. not soft like 'what was your most difficult project' but something which required a precise answer.) These may be something you've been asked or asked as an interviewer. Overall, such information may help me figure out if I am a fit to STEM fields at all. My preference is data analytics/computer science followed by all kinds of engineering.

Thanks,

Monte

2. Apr 21, 2013

AlephZero

So you're not smart enough to figure out that interviewers also read books? OK, we'll call you back (or maybe not...)

What you know may be less important than how you think, especially "thinking on your feet". In that case, your strategy won't work. If a candidate "knows the answer", we just keep changing the question till we find something he/she doesn't know about. And if we are feeling vindictive (or the candidate seems really good and we want to find out just HOW good), we might disagree with whatever he/she says, even if it's correct. (We know plenty of wrong answers to the interview questions - we learned some of them from previous interviewees!)

Last edited: Apr 21, 2013
3. Apr 21, 2013

Monte_Carlo

Questions/Problems

Can you mention some specific questions/problems/puzzles to get a feel for difficulty levels? Anything concrete would do.

4. Apr 21, 2013

wolfmax

1) How do mobile phone calls work? What can effect them?

2) If bikes had all there weight distributed on the circumference of the wheel what would happen?

3) If all cars in the world switched to hydrogen power, would the water produced raise sea levels significantly?

Hope these help

5. Apr 21, 2013

Monte_Carlo

Concrete Interview Questions

Wolfmax, looks good, what are the answers to 2 and 3? I would appreciate more visitors posting similar questions - I know I should ideally consult literature but it's of no use, I better hear it straight from the source.

6. Apr 21, 2013

wolfmax

With the bike it would be harder to steer due to larger angular momentum. It would also be harder to pedal. With the hydrogen car one it was more a test of your estimation skills. The answer is not important, it's to let the interviewer see how you try to figure something out and if you have a feel for what you need to know to work it out. Also it probably weeds out people who stare blankly for 10s and say 'I dunno'

7. Apr 21, 2013

Staff Emeritus
Actually, since the hydrogen comes from electrolysis of seawater, it cannot raise sea levels. Indeed, it will slightly lower them.

It is a sad but true fact that there are college graduates who are technically inept and/or incapable of effective thinking and/or communicating. The point of these questions is to try and not hire them.

8. Apr 21, 2013

bhobba

I worked as a programmer for 30 years before retiring and interviewed countless people.

You bet interviewers ask questions that make you think on your feet - especially those that demonstrate understanding - not rote learning.

My favorite one was most software projects actually fail to deliver on time and on budget and its generally considered to virtually never be the fault of lack of technical expertise but rather failure of methodology and/or management. Why do you think that is? And what would you do about it?

Now the cause is a matter of opinion and the solution to the problem just as much an opinion - the specific answer is not the point of the question, but rather do you understand the issues involved. My preferred solution personally was a RAD many steps often approach where we pull back and analyse the outcome of each step before proceeding. But other approaches most certainly are used - its not the particular approach that is of interest to the interviewer - but rather do you understand the issues involved.

Another issue to consider is the level of the job you are going for. Generally if its a base level job questions at a detailed technical level are asked but as you go up the ladder you will find management types start to appear on the interview panels and those types of questions also get asked. On one I was on for a team leader position which is what I was my director was also on the panel and she of course has a different focus. She asked - what are the types of things you can manage. The blank looks and fumbling that we got to that one was unbelievably shocking - I just couldn't believe it. As a team leader you only really are involved in a bit of management but you should have some general knowledge.

Thanks
Bill

Last edited: Apr 21, 2013
9. Apr 22, 2013

carlgrace

When I interview someone I ask about a project listed on the resume. After the candidate explains it, I drill down and ask questions, deeper and deeper, until the candidate doesn't know the answer (I usually won't know either, but I can tell if the candidate is hand-waving). Then you get to see what the candidate is made of, and only then.

10. Apr 22, 2013

bhobba

That's what I thought as well - and that's not my background.

Actually my experience in the area of software engineering is most graduates are technically sound but what they lack is the ability to think with that knowledge and apply it beyond what they learned. Its something they pick up with some job experience. It always amuses me when I see posts of people wanting to get into some ultra elite and prestigious program thinking the knowledge they gain in such programs is what will give them the edge. That's not it at all - the edge is in being able to think with what you do know not the amount you actually know.

That's what the interview questions I, and others I know, were aimed at. Fortunately for me my interview was a joke - they simply asked rote questions with rote answers from a software engineering textbook - back in those days there was massive short supply of programmers so interviewers were looking for any way they could give someone a job. Unfortunately its not like that now.

Thanks
Bill

11. Apr 22, 2013

Devils

Most hydrogen in industry (the hydrogen economy) comes from hydrocarbons.
Electrolysis accounts for 4%.

http://en.wikipedia.org/wiki/Hydrog...ydrogen_market_.28current_hydrogen_economy.29

12. Apr 22, 2013

Staff Emeritus
Today. But if you put every car on hydrogen (the premise), you don't have enough fossil hydrogen to do it. Apart from the political issue of replacing one fossil fuel with another. But the point of questions like this is not whether you can get the right answer on your feet: it's to show you can think on your feet, can consider issues, and can prouce a coherent answer.

13. Apr 22, 2013

bhobba

Thanks
Bill

14. Apr 22, 2013

Devils

You weren't this argumentative when you used to read Omni Magazine & Douglas Hofstadter.

15. Apr 22, 2013

bhobba

Gee - that must have really been in my past over 30 years ago when I was doing my degree part time and worked as a clerk in a government department. Since then a LOT has passed under the bridge including exactly how technical team structures actually work. I rose pretty quickly from base level programmer, to senior programmer to team leader but stalled there because I was judged to not really have any management expertise.

That's the real key to job advancement - not technical expertise.

Thanks
Bill

16. Apr 22, 2013

Devils

Well back in those days, "operators" didn't have degrees & programmers knew how to program. Now I see "help desk specialists" with masters degrees, and people with IT degrees who never program. With so many people getting degrees these days, you almost need one to make the coffee.

I would never recommend IT now, it has no barriers to entry. Instead people are going into nursing, engineering, teaching, even trades like carpenter or plumber. You need a formal qualification for these, and the qualifications are readily transferable internationally.

Probably the best skills you can have in life are likeability (plenty of psych literature on it) & communication. High IQ like most have here may get you to some low paid postdoc position at age 30.

Success is not well correlated with intelligence; its more correlated with persistence. If you read "The millionaire next door", the meek guy who is a cleaning contractor with 10 employees, making $200K & driving a 10yo car, could be far better off than some stockbroker earning$500K and spending it on status symbols.

And yes you did cook a nice steak.

Last edited: Apr 22, 2013
17. Apr 22, 2013

bhobba

Oh dear - too true - too true.

Yea. But it varies. The first place I worked at after graduating, the Federal Police, I was VERY well liked and got along with nearly everyone. Then I moved elsewhere - and it was mixed - some just loved me (strangely they were mostly business people - my director then constantly gave me feedback they just loved me and I was the only programmer she ever came across like that - mostly business people think technical types are elitist and think they are 'above' non techies), others thought I was so so, still others including the director I had before retiring (not the director referred to earlier) thought I was a total idiot.

Since this is a career advice section those just starting out take note - find a place where you are well liked - that's much more important than anything else. I absolutely regret leaving the Federal Police chasing bigger bucks and better career prospects. The most important thing by far is what others think about you.

One of my favorite books - read it years ago - should have taken it to heart more.

You must have known me just after I graduated - I didn't do much cooking before that - I stayed with my parents.

Thanks
Bill

18. Apr 22, 2013

JakeBrodskyPE

When you march in to the working world, a funny thing happens. Nobody gives a damn about theories. They care about results. A theory is only a model of reality, not the reality itself. This is not about doing some stupid word problem; this is about coming up with answers that are achievable and that make sense.

So what did you sweat all that time in school for studying theory? You study it in the hope that you'll find the concepts useful. And here's the rub: most of what you learn in university you'll never use again. You'll use maybe 10% of what you learned. We don't know which 10% it will be, but we know that it is unlikely that you'll use more than 10% of what you learned as an undergraduate. And with each degree after that, the figure drops in half.

What I'm trying to convey is that Interview questions are designed to see how well you reason, how pragmatic you are, and what you make of a truly unusual situation. Why? Because that's what you'll see from now on.

The mentality that most learn in school is that there is usually a right answer and that's it (unless you're dealing with roots of an equation). In the real world there are a multitude of answers, and we're trying to figure out how you'll wallow your way through them all, and how you reason toward an answer that is close enough to get the job done without undue effort.

This is why I always hated word problems. You might get answers, but the questions usually sucked.

19. Apr 22, 2013

jesse73

Dont studies show that individually being a jerk is the best strategy because you are perceived as more confident and competent. On the other hand the success of the organization as a whole is actually undermined by the same incentive to be a jerk mechanism for individuals.

20. Apr 22, 2013

Monte_Carlo

Specific Questions

Bhobba,

Could you provide specific examples? I love to hear actual programming interview questions which make you think intensely. If you could recall a few, that would be nice. Also, if you could recall how much time was given to a candidate, that would be awesome. But remember - these don't have to be programming questions, they could be generally algorithmic so applicant won't have to know language in detail.

Monte