Interviewing New Graduates 8

I give 'em a few questions that they should be able to figure out, like:
- Sketch a simple beam, call it a bridge across a creek, and ask what they need to know to figure out if the resident alligator can bite them as they walk over. Not asking for a specific formula for deflection, just asking for what parameters must be known in order to solve such a problem.
- Ask a little about the metallurgy of steel, like what makes it different from iron, and how much of that something makes a difference.
- Ask about preloaded bolts and fatigue.
- Ask specific questions about things they claim to understand; only the ones you actually understand, of course.

IMPORTANT: Do NOT answer the questions or reveal whether the candidate was correct or not. Just note their answer and keep going. The intent here is to not educate the candidate during the interview, to not waste your limited time, and also to apply a little pressure, the same sort of pressure they will feel every day at work, when they are given problems for which the answer is not known by anyone beforehand, and the correctness of which cannot be established right away.

After the interview, find a quiet place and write a summary of what was said, and how the candidate reacted. Include the candidate's name and the date. Candidates who know their stuff will either not be bothered, or will ask questions about what you are trying to figure out about them. The ones who are just winging it will either seem oblivious, or will sweat profusely.

Keep your notes and summary in your possession, preferably at home.
Do not forward them to HR to accompany the inevitable form.





Mike Halloran
Pembroke Pines, FL, USA

One of the more important jobs of engineers is how they present their data and recommendations. How much do they know about computers and what is their grasp of the English language. What are their outside interests and how do they get along with others. Is this a person that some day will be grinding out work or will they be a supervisor? I've also found t that having some mechanical ability and experience outside in muddy and bad weather helps in the early years.

As a geotech, I like to see that a college grad took a geotech elective class (even better if they took two). This helps to weed out those interested in my field and those just looking for a job. It also gives them a stronger fundamental base to work from. I also ask why they want to do geotech. They won't fully know why, of course, but you can still get a feel from their answer.

Internship experience is good, but like you said, not necessarily a slam dunk. I do like to see real world work experience of any kind. Even working in a restaurant for more than a year shows maturity. Kids who worked part time in high school probably have a good work ethic. I ask technical questions, but I don't expect them to answer it correctly. I'm more interested in hearing how they process the problem. I do share the answer with my candidates because it engages them. Remember, you are not just picking them, they are picking you. If you want a good candidate to pick you, then you need to show them that they can learn from you.

Mostly, just asking open ended questions to get them talking is the best way to interview. Don't lead with your questions or you'll get canned answers. Questions like "do you mind working long hours to meet a due date?" are the dumbest kinds of questions.

It's good to have a list of questions prepared and ask everyone the same (or at least mostly the same) questions so you are getting a less biased comparison. There will be some gut involved and you will make bad picks given enough opportunities to hire people. That's okay.

One thing to poke at are claims of experience on things or software. I once asked a candidate what systems they had modeled in the software that they claimed to be familiar with. Their response was that they "watched" someone else use the software, and hadn't ever used the software in question.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list

Review the resume closely for lies, omissions and exaggeration etc. anything fishy or doesn't pass the snuff test then question them on it. Often Software falls in this category as IRstuff alludes. I once noticed the claimed number of hours experience on a specific CAD SW on someone's resume looked high (not a new grad) so I did the math. Basically the guy was claiming 40 hrs per week using the SW - no vacation or holidays off, no time in meetings or other activity, no sick and no on the clock toilet or coffee breaks etc. - from the day he graduated. Speaks to him either purposely trying to mislead me or not thinking thru something as important as what he put on his resume- I extrapolated that to what he'd be like to work with and did not recommend him.

I also back up the language/communication skills oldest guy mentions. We have a lot of non native speaking applicants historically and I used to try not to be too harsh in assessing their English skills. However, the ones that got hired turns out that those with better English skills turned out to be better employees overall and some with weaker skills really didn't work out at all. So now it's almost a deal breaker if I have any communication skills at the interview. Obviously there will be exceptions but for my kind of work few and far between IMHO.

I'm not big on the technical questions. I'm hoping that anyone who got or is getting a degree can do the simple problems we can ask in an interview. And if they can't, maybe they don't work good under the pressure of having some old guy toss confusing, poorly presented problems at them. I prefer showing them what our work involves and seeing what questions and input they have. I remember at my first interview, the assistant division head asked what I would space the columns in a building. (By the way, they didn't design high rises)I looked around the office and guessed 40 ft. Wrong, but I got the job. I lasted there 20 years, so I guess I did all right.
I'm much more interested in sizing up their work habits. Will they show up Monday through Friday at a set time, basically for 40 years. Are they curious? Are they interested in actually doing engineering? Will they work well with others? Are they going to bring a lot of drama to work? Can they travel?
If they learned the basics in college, I can hopefully fill in the gaps.

I think I would mainly just suss out if they like engineering. Most stuff can be self taught or learned and most engineering jobs for the most part aren't as technical as we would like to pretend them to be. Not being able to write and be understood would be a deal breaker with me.

"I'm hoping that anyone who got or is getting a degree can do the simple problems we can ask in an interview. "

Not always. We had a candidate who supposedly got a 4.0 GPA in college, but couldn't solve a freshman EE problem, not once, not twice, but three times during the day, AFTER getting the solution the first two times. That was sad, but sadder was the fact that was the best question the 3 of us could come up with ;-)

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! faq731-376 forum1529 Entire Forum list

I always find it interesting to see if you can get a new grad to say those three hardest words:

"I don't know"

I'll also second the value of working during school, and being able to articulate why you like engineering.

----
The name is a long story -- just call me Lo.

I don't interview a new grad to find out what someone knows. I interview to find out how they think.

Who is right doesn't matter. What is right is all that matters.

As others have said you need to find out how they think and solve problems. Just ask them to explain how they would approach it and what tools they would use.

What do you expect them to know from school? Formulas, material properties, terminology?

I have always paid close attention to what questions they ask. I find that most interesting in any interview. How do they react to a shop tour or when you show them your products or drawings?

Honda is one company I know has written tests. There are others.

This difficulty in demonstrating what a student knows to someone in industry is why I always recommend students work on design projects or competition teams.

I haven't been back in industry long (I've been teaching college for a while, and yes, I emphasized design skills), but I'll be asking two questions.

I will ask them to describe the engineering design process, then explain how they applied that process to solve a design problem at school, including discussion of constraints. Recent grads are typically well-prepared to answer engineering analysis questions, but clueless when it comes to design. Those that can discuss design will get my approval. Those that show they don't know what design is by describing analysis and explaining how they solved an analysis problem don't get my approval.

Since I'll be hiring electrical engineers, I've decided I will ask one multi-part technical question related to electrical design. I've tested this question on many classes full of students, so the results I'm discussing below refer to their behavior.
[ol a]
[li]You have been tasked to power a 5 V ± 10%, 1 A resistive load from a 10 V supply. You've been told to use a voltage divider to create the 5 V. Select resistors for use in the voltage divider.[/li]
[li]Determine the efficiency of your design. Propose alternatives that will have a better efficiency.[/li]
[/ol] For Part a, most of my students would know the resistors should be the same resistance value, but then they'd get stuck. They didn't know how to choose a value for the resistance. Most threw up their hands at this point and said "I don't know how to continue."

Once in a while, someone would arbitrarily proposes a resistance value (often it was me - one I knew to be incorrect), then we'd analyze the circuit to see if it would meet the specified voltage to the load. Usually, their design didn't work, so I asked them what they wanted to do next.

Someone who understood the design process would suggest we iterate (I often had to bring up this idea, but occasionally it was a student), choosing a different resistance value, and analyzing again. They'd usually go in the wrong direction with their next proposed value. Since now everyone in class saw the need to iterate, they'd iterate again and go in the opposite direction from their first iteration, and after a few cycles, they'd get a workable resistance value.

Once they'd get a resistance value that works, I asked them if we were done. Only a select few knew they need to spec a wattage rating also.

It would be pretty straightforward analysis at this point to calculate the efficiency for Part b. My students could usually do this.

I imagine, like my students, most recent grad interviewees won't do well with this, so I'll move them on to Part b at any point that they give up. My plan is to give them some resistance and wattage values that will work to get the voltage within tolerance and have them show that the values work using analysis. Then they'll have to calculate the efficiency of the design I've given then - it's horribly low. I will finally ask them what more efficient alternatives they would propose and how they would push back against management's direction to use a voltage divider. This will let me judge some of their soft skills.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

I've been doing this for a decade. A few tips:

1) Interviewing is a crapshoot. After a decade doing it, I'm better at selecting people via an interview than I would be picking the resumes of my pre-screened candidates out of a hat at random, but I'm not sure by how much. Sure, I can weed out the total dolts, but I'm not 100% confident that I'm selecting the very best candidates either. So treat it as a pre-screening tool only, similar to reading and reviewing interview packages (which include a resume and a transcript for a fresh grad)
2) Ask technical questions. That will rapidly allow you to sort out the people who merely have good marks from the ones who have real conceptual access to their education. You can test for "practicality" with the right questions, and find the people who have an understanding which is informed by experience with the real world, if that's important to your business. It definitely is important to ours!
3) Hire them for a paid internship with a definite end date, preferably short- ours are 4 months. The problem with 1) is greatly mitigated when you're just using the unreliable interview process to select candidates for what amounts to a 4 month paid interview. We don't like 8 month internships and hate 16 month "experience year" internships because the selected people inevitably become part of the furniture, and can make it on to staff for reasons other than their abilities or potential as engineers.
4) If you like the student, and they like you, hire them back for a 2nd four month interview, after more schooling. After eight months of interviews, you'll have a very good idea who they are as people in a work environment, which you can only get the vaguest sense of in an interview.
5) Select the very best from your former interns, and hire them on staff. The customary contract period at the outset isn't really required at this point- they've been on contract with you for four or eight months already.
6) Invest some effort in training them on the job, with proper mentorship.
7) Pay them fairly and give them interesting work.

That's been our formula. It has worked like a charm. We have a far higher failure rate with staff who were hired after nothing but interviews, even when we used 3rd parties to do psych evaluations etc. Our former interns almost invariably turn out to be excellent employees. And if you do 6) and 7), turnover is quite low. You do get those who leave for family reasons etc., but that's inevitable.

I have found that it is not necessarily what he or she has learned in school that counts, its that they have the ability and willingness to learn what school did not teach them. College does not always teach you what you need to know, it teaches you how to learn.

Ask how many millimeters are in one inch. (I am serious - you may be surprised)

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

CheckerHater said:

Ask how many millimeters are in one inch. (I am serious - you may be surprised)

Click to expand...

A LittleInch or a BigInch ?

Metric Inch

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

I used to work for a company that did business all over the world, in inches and in mm.
We were constantly converting/scaling CAD files back and forth.
For a while there, we were making parts that didn't fit.
... because one of our experienced fabricators was subbing as a designer,
and using 25 mm := 1 inch for his conversions.
It took months to undo the damage.





Mike Halloran
Pembroke Pines, FL, USA

My favorite is when people translate a CAD file (can be done 2D or 3D) and mismatch the units so it comes out a factor of 25.4 off. Customers are the worst for it when we send them floor plans for setting up some of our big instruments.