Interviewing candidates 22

[slickdeals]

I am seeking feedback on how some of your firms conduct interviews for engineers in the 0-3 years of experience. More specifically, are your interviews strictly non-technical in nature or a combination of non-technical with a 10-20 question test to check their technical abilities.

I would like to introduce a technical aspect to the interview, nothing too fancy with higher level math or fancy engineering, but checking for common sense and foundational principles key to succeeding in our business.

Thanks for contributing!

 

[bookowski]

Anyone care to share any actual interview problems? I'm interviewing a guy with 2.5 years of experience in a few days. I tried an exam once in the past but way overshot it, what I thought was a 30 minute exam was apparently hours.

The ability to visualize a problem from a description seems to me to be one of the key indicators of proficiency. Everyone I know that's worth their salt can digest my half assed explanation of a problem and start opining. So I was thinking of typing out some structural scenarios and ask the interviewee to sketch it. I may be overestimating again though.

 

[slickdeals]

If we come up with a set of interview questions, pretty soon this post will make it to the top of a Google search I have some questions formulated, mostly all sketches with questions.
1. Combined footing with a question on what the moment diagram would look like and where you could place rebar.
2. Retaining wall and basement wall sketches requiring moment/shear diagram and rebar.
3. Drawing a moment frame with lateral loads and asking to pick which one represents shear/moment diagram. Following on, showing them a sketch of the beam with top bar with properly developed hook and bottom bar with 6" embedment and questioning what's off.
4. Drawing a beam supported on 3 springs, the middle one being a soft spring and asking them hypothetically how the reactions would be, which would have higher reactions etc. More to see how they understand load follows stiffness concept.
5. Drawing a simple 2D shear wall on a GB with one pile each end + an axial load and moment and asking to calculate the pile reactions.

Stuff like this is what I was envisioning for a entry-level engineer (0-3 year). What say you folks?

 

[milkshakelake]

@bookowski I can't share mine because it's not entirely owned by me (developed with another structural business owner, which was a good idea), but you can always adjust the questions if you have issues with them. It takes refinement over time, like with calculation standards. I think all the comments above pretty much summed up what to look for. You can find random questions online and tailor them to exactly what you need and are looking for. For me, a big part of it is just finding out if they can follow instructions/training and if their ego will get in the way.

For technical questions, it goes from easy to hard. I have a cantilever moment question (something like 100 lb horizontal load at the top of a 10' column), basic moment diagram (the famous wl[sup]2[/sup]/8), axial stress thing, soil bearing stress thing, frame analysis thing, sketch a concrete footing of any kind, and then some moderately hard to insane questions about seismic, shear walls, collectors, etc. The whole thing can be done in 20-30 minutes, or maybe 5 minutes if you're KootK. A shockingly large amount of interviewees who graduated with a civil engineering degree will get 0 of them correct, stating that they forgot it because they learned all this stuff years ago, at which point the rest of the interview is just a quick formality to say goodbye.

 

[Celt83]

I'm not involved in the interview process but from current mentoring experience and how folks work today I would tee up a handful of software result interpretation-based questions, mostly a tell me what is wrong (or what the software did is exactly right but wrong for the design intent) with this picture and why type of deal and build into these questions your typical draw me the bending moment diagram/deflected shape questions.

I'd also be looking for someone to tell me point blank when they don't know an answer and even better get a "I'm not sure how to do this but I'd really like to know can you walk me through it?"

 

[WesternJeb]

As a younger engineer, your idea of a screening exam scares me if you are proposing all of those at once.. While the questions aren't overly difficult, those are still some of the easier PE exam depth level questions. I would advise maybe only 2-3 of those questions and use them as a conversation piece to determine their teach-ability, as others have suggests.

I would like to second the post above with determining computer skills.. We had an intern this summer who could barely type beyond pecking; we had assumed any younger engineer likely had computer skills but learned that lesson.

 

[WARose]

As a younger engineer, your idea of a screening exam scares me if you are proposing all of those at once.

Agreed. I'm a senior engineer, and if somebody threw all that at me....I'd be thinking about other opportunities.

 

[bookowski]

4. Drawing a beam supported on 3 springs, the middle one being a soft spring and asking them hypothetically how the reactions would be, which would have higher reactions etc. More to see how they understand load follows stiffness concept.

@slickdeals - I had almost this exact question on my previous exam. I had a 3 span "very stiff" continuous beam, so 4 supports. A vertical point load at the middle. There were two scenarios, one where it was supported on "very soft" springs and the other where it was supported on pins. The question asked to draw the reactions - direction and relative magnitude only (no numbers, just using big and little arrows) for the two conditions. I thought it was maybe a 6 out of 10 on difficulty but most people were stumped.

 

[Aesur]

I'm not a huge fan of crazy exams for interviews, however I think some very basic skills can tell a lot. Try having them measure a few things on some plans (can they read a scale), give them a detail missing something, ie truss on top of wall, but don't give any connections, ask them to sketch in the connections (ie, uplift and out of plane attachment). Maybe give them a roof plan with slopes and have them sketch in a general roof framing of trusses, give them overhangs or a corner condition. Don't make them run any calculations or even need to open a book to look something up, that is a waste of time and can cause a good candidate to freeze or even question why they are interviewing there as they may get a ill taste for the company early on and there is no coming back from that.

I find if a person can understand a detail they more than likely can do the engineering.

 

[Boiler106]

I don't doubt it. But, then, we can only work with what you've chosen to share. And, on the subject of whether or not technical exams have validity, what you've shared is that your organization seems to have decided that technical exams are not meaningful based on the statistical significance of singular data point.

thank you for insulting me through a condescending bluebeam sketch. that's a first for me. its simply impressive how much effort you put into convincing a faceless stranger that you're right.

 

[KootK]

thank you for insulting me through a condescending Bluebeam sketch.

I apologize for insulting you Boiler106. I've clearly misread the room on this one. We are most definitely not having fun here it seems.

its simply impressive how much effort you put into it.

I felt obligated to clarify things given that I'd botched the original description as you yourself pointed out.

 

[Boiler106]

kootk, apology accepted, however, our decision was based on the results of a much larger sample size than one kid (a kid that broke the camels back, as it were) coupled with our realization that other qualities ranked higher for us. we all write for the ncees exams and are very familiar with testing.

 

[YoungGunner]

We mostly hire students as fresh hires and then use the following years to pick the cream of the crop as salaried, but there was a time where I had consecutively extended 25 offers and had 100% acceptance rate. This is my interview process. Keep in mind we even hire freshman and sophomores who have never had a structural class yet:

1) To get to know them, I start of by saying I'll introduce myself in 30 seconds and then ask them to do the same. I do it first so I can establish who I am in the company and as a person (I mention a hobby or two) and then they have a template of what to say.
2) My first question - why did you become a civil engineer. My purpose is to see if they talk about something unrelated to structures. I don't want those who are interested in water.
3) Second question - what interest do you have in our company? (or something like that. The purpose is again to see if they have any other motives and give a chance to explain themselves)
4) I then ask any questions from their resume, just to give them a chance to talk more.
5) We go over the benefits of working for us (performance reviews and raises every 6 months, 30 other students that are potential contacts after you graduate, etc)
6) I then turn it to them if they have any questions. I'm a bit disappointed if they don't have any, or the only question is "how much would I be paid."
7) I then give my non-technical evaluation (shared here for your viewing pleasure). I preface that this is just a way of seeing how they problem solve. The point of the questions is this:
1. See how they attempt to mark the center.
2. See if they read through the instructions first or just go go go. Annoying when engineers read one line and then GO without reading an entire email.
3. This one trips up about 50%. I don't mind if they get it wrong, but I want an explanation other than "there was no question to answer" or "I don't know what it is."
4. This one is all about problem solving. The most common way of solving it is to write their name on the ripped piece of paper and put it below. I again don't like it when candidates give the reasoning of
"it's impossible" or "I went with the first instruction." Too many times in engineering where we get conflicting information and "I did nothing" is what students do.
5. If they don't get this one, it's an immediate no for me. 90% get this right and I see a pattern with those that don't get it right.
6. I'm surprised most don't catch the "10 words or less" in regards to the provided response. Most rewrite their own because they don't like what it originally says.
7. I recently changed this one and I'm not sure how I feel about it.
8. First of all, I don't actually ask them to bring a pink marker in the email. But I have a pink marker next to me during the interview. So the goal is two fold - see if they have the courage to ask for
help when they may have messed up, and then see what they do. Some draw their own box to the right, some do nothing (which again I don't like), and the correct answer is to highlight the box above that's to
the right.
9. This question is partially to conceal the answer for #3 but also to see if they simply trace one of those double lines. Few do, so I've learned to just see whether they follow the instructions. Had a few
people ask for a paper or a ruler or pull out a book, though it says not to. I had one person draw a clearly terrible straight line and comment "I think I did a good job."
8) I then give them a chance to explain their reasoning behind each problem.
9) Now this the moment I sell them or reject them. If I'm unimpressed with everything up to this point (them as a person, the evaluation) then I end the interview. If I'm impressed, I ask if they'd like to walk around the office and see what's being worked on. I then bring them to our open area and interrupt some of our students in the office and ask what they're working on. It's a great moment for the candidate to hear from a student what they do, and see how much there is to learn. They get to ask the students questions. I do this for about 2-3 students before asking the candidate if they have any final questions, after which I tell them I'll let them know THE FOLLOWING WEEK.

The whole interview, including walking around the office and evaluation, is only 30-45 minutes. I show them clearly what the benefit is of working for us, and I got to see them in various settings and how they handled themselves. It's also a moment where one of our students may recognize the candidate, and then I can ask for insight afterwards. I don't do multiple interviews - I agree with those that you should be able to figure it out in one go.

 

[271828]

If we come up with a set of interview questions, pretty soon this post will make it to the top of a Google search smile I have some questions formulated, mostly all sketches with questions.
1. Combined footing with a question on what the moment diagram would look like and where you could place rebar.
2. Retaining wall and basement wall sketches requiring moment/shear diagram and rebar.
3. Drawing a moment frame with lateral loads and asking to pick which one represents shear/moment diagram. Following on, showing them a sketch of the beam with top bar with properly developed hook and bottom bar with 6" embedment and questioning what's off.
4. Drawing a beam supported on 3 springs, the middle one being a soft spring and asking them hypothetically how the reactions would be, which would have higher reactions etc. More to see how they understand load follows stiffness concept.
5. Drawing a simple 2D shear wall on a GB with one pile each end + an axial load and moment and asking to calculate the pile reactions.

Stuff like this is what I was envisioning for a entry-level engineer (0-3 year). What say you folks?

For a new grad, depending on what courses the interviewee had, some or maybe most of those might be unrealistic. The interviewee might've never seen some of those scenarios/components in any class, even at a decent state school.

If I was going to set up an exam, I'd look at the transcript first, and ask some questions about the structural classes they had. If they had a class in it, and still couldn't answer, then that would be a fair test.

Also, a question like #5 is beyond the expected judgment of a new grad, IMO. If they had a foundations class with piles, they would think you described a highly statically indeterminate problem (which it is) and wouldn't know that they could simplify/idealize and work it using statics.

 

[milkshakelake]

I thought it was maybe a 6 out of 10 on difficulty but most people were stumped.

I don't know if eng-tips naturally selects smart engineers simply by virtue of existing (i.e. you have to be able to use a computer and ask a question), but 90% of the structural engineers I know in person probably couldn't answer any of the posts here. So for new graduates, you really have to tone down the difficulty. I think your question is very reasonable for an experienced engineer, maybe 4/10 instead of 6/10. But for a graduate, that's 10/10. They wouldn't know the difference between a pin and a spring.

 

[Enable]

Take this post with a requisite grain of salt as I'm someone who has never interviewed for a position in the sense being discussed. None the less, I am determined to persevere by providing my Canadian two cents. I’m also on the hiring side of the equation, so that should make me eligible to waste a few keystrokes.

If technical exams were more than marginally informative about a fresh graduate’s aptitude their transcript would be all you need. After all it already tells you about their proficiency with these kinds of tests (lots of them). Better yet, their results would be a product of a more realistic environment compared to the pressure cooker of an interview with all you sexy son of a gun engineers! But the very reason we are talking about administering a technical exam during an interview is that most (all?) seem to agree that grades in coursework don’t provide enough information to make an informed decision regarding candidacy. So, what are we advocating here? That a slew of university exam results is close to useless but if we make and administer a single (short) exam ourselves, we would get a heck of a lot more mileage out of it? That seems kind of hubristic of us, don’t you think?

Of course, if it was harmless then there would be no downside to doing so. It would just be a waste of time. But it’s not useless. As others have mentioned, an ill-conceived exam could be construed as something sinister (or whatever) and cause future graduates from a certain school to shy away from your firm (people talk and all that, and when jobs are bountiful fresh grads have options). This is undoubtedly a larger issue for smaller firms than the big boys, but I think that includes most of us in this thread. To me it seems like technical exams might be more useful for an engineer who has purported technical expertise and has been out of university for some time. If they can’t do some of the things slickdeals mentioned fairly rapidly, then god help their designs. For a fresh or recent graduate, there have to be better ways.

I am extremely excited about the interview process mentioned above about having the candidate bring in a project to discuss. That allows for two-way discussion that seems more likely to reveal what we actually want to know: does the candidate’s mind work right or are they just able to shift symbols around the page? It’s great because it also is something they are familiar with, which should reduce anxiety and thus increase validity of our impression of the discourse. Best of all I like it because it allows the candidate to evaluate the firm! An employment relationship is a two-way street. And I would rather have prospective employees be able to figure out if they think that they would be happy / a good fit (for themselves) during the interview than to be hired and hate their life; creates high turn-over, which costs money and hurts morale. I think that the project thing would really go a long way in helping facilitate this.

Last thing that is somewhat tangential. In my spare time I am a non-engineering professor at technical college, and I also teach a prep course for the law school admissions test. I can assure you that it is a LOT harder to write clear, and concise questions that provide all relevant information (with minimal ambiguity) than you might think. Don’t underestimate your own shittyness at developing exams. If you ever want to take a whirl down this road for shits and gigs compare an average LSAT LR section to a GMAT CR section; it’s almost like toddlers wrote the GMAT CR section by comparison. That’s not to throw mud at the GMAT, most other standardized exams I've reviewed seem to be in the same boat. It’s just really freaking hard to write good questions and the psychometricians at LSAC have done a fine, fine job.

 

[wrantler]

My 2 cents. I think it is OK to have technical questions or methodology questions. But I would think those are more to 'weed out' the ones who may have gotten to the top of your list by accident. Almost like a 'pass/fail' type of scenario. I don't think you should rank them on how well someone answered a one off question. I had a colleague who somehow memorized the codes and would randomly quiz staff. What an idiot he looked like when doing that - nobody cares that you memorized the code word for word.. He quit as the pressure to actually make engineering judgment decisions got to him.

After all we are a business. So you need to have a 'sense' to be effective on projects as many have said before me. I think this is huge. If you hire an overly technical person who calcs every single thing that could be checked under the sun....and still cant make a decision they are not a good practising engineer.

 

[SwinnyGG]

We mostly hire students as fresh hires.....................I agree with those that you should be able to figure it out in one go.

This..... is wild.

If it's working for you, my opinion is no reason to stop doing it.. but I bet you're missing a lot of future excellent engineers because they are nervous and you're deliberately trying to confuse them. I'm a successful engineer with a lot of experience, and I would've failed your test handily.

To each their own I guess.

 

[dik]

Why? The only thing puzzling was whether I shaded the math answer pink, instead of the box on the left. Math question was a piece of cake seeing the 0 outside the brackets in the numerator. The sentence as a bit of a puzzle, no matter how I looked at it, it didn't make much sense, so I suspect the answer wouldn't have to make much sense. Exact centre determined either visually (people are generally pretty good at binary division, except after a few beer), else fold the paper in half and give it a bit of a pinch in the middle (X marks the spot).

I forgot to include, the answer to the Civil Engineer question, being far to broad for 10 words. My answer, "Historically, civil engineering is the other type of engineering."... 9 words. This reflects the broad scope of engineering. Historically, there were two types of engineers, Civil and military. Civil was all encompassing.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[phamENG]

YoungGunner - I have mixed feelings about your test. As SwinnyGG mentioned...it's a bit rough. But on the other hand, I know what it takes to be a 'graduate engineer' in a structural consulting firm. If you build your entire firm out of young, career driven, unmarried 20 something students clever enough to pass this, you're either on a rocket ship to the moon or a burning bus driving off a cliff...I'm not sure which yet...

 

[geesaman.d]

If technical exams were more than marginally informative about a fresh graduate’s aptitude their transcript would be all you need.

I was thinking of a friend of mine who I studied with extensively in university and we both worked our first jobs at the same company. He's a damn good engineer (not sure if he's still doing it) but not a good test taker. He got through engineering school with good grades but in senior design and as a design engineer he danced circles around the other new grads. Why? Because he understood the test matter, had good (but not outstanding) math skills, worked hard, was detail oriented and organized, was a great communicator, put in extra hours when it mattered, saw the business context behind every transaction, and happily collaborated with more senior engineers without wasting their time. I think about my company's interview process and I think he would have done well. He would have been miserable about the test and probably not want to work there because actual engineering is not answering test questions.

I agree that if standardized tests are only marginally effective at finding the best engineers, then little quizzes put together by an individual will be no better than that. This may offend some individuals who can consider test taking as one of their strengths. But it really does have limited applicability to real world engineering capability.

There is not one perfect engineer for all jobs. I suggest building a list of the best qualities and sort them by your organization's needs. Then consider how the interview will explore each of those points and how much time will be allotted. (I'm a fan of discussing a technical project or problem solution because it hits several at once.) If being able to pass the EIT and PE exam is critical, perhaps borrow a couple of those questions. Otherwise I believe that taking a quiz or test does not make the cut for most organizations.