T-SQL Tuesday #174 Round-Up

Welcome to this round-up for T-SQL Tuesday.

The topic for this month was, What is your favorite job interview question? In that introductory post, I made it clear that interviewers and interviewees both can and should have questions, and we received a nice admixture of the two sets from the lovely contributors who, well, contributed.

Our Interview Panel

I thought about doing this in the normal Curated SQL style, where I grab a graf from each post. But instead, you get the second-best Curated SQL format: the hodge-podge bulleted list, but still in present tense and with a touch of commentary and the occasional rant.

• Rob Farley wants to know who you are. This includes understanding the technical side of things as well as the personal side.
• Kevin Chant drills in until you don’t know. Having to admit, “I don’t know the answer to that, but I’d learn the answer by {doing thing}” is absolutely fine in an interview, assuming you don’t hit the wall too early on. Kevin also has good advice for interviewers.
• Reitse Eskens discussed his company’s interviewing experience. Finding the right culture fit and then performing a skills test was interesting—generally, my experience has been that companies try to winnow down on skill level and then select for culture fit. I suppose my question is, which filter is more selective? And in fairness, we could also discuss whether that’s the most important question.
• Mala Mahadevan pokes her head out from the books and includes a variety of questions from both sides of the table. Mala includes several “Please don’t ask this because it’s a really dumb question” types of questions. As for the correct number of deadlocks, the number is 462. 462 or fewer, no problem. 463? Burn the thing down trying to fix the problem. The intelligent follow-up question is, “463 over what time period?” But if you have an intelligent follow-up, why start with an unintelligent question?
• Andy Levy asks a great question and not just because he mentions PolyBase. It’s useful to keep up to date on SQL Server, not just in terms of shiny new features but also on T-SQL changes that can make your life better.
• Andy Yun wants to know that you know how to learn. And that you don’t fall into the most common habit of the person two chapters ahead in the book: casual arrogance. No, mine is a studied and practiced arrogance and I attend regular arrogance seminars in order to learn the newest techniques around how to act abrasive around people.
• Erik Darling slips the bouncer a $20 to get in, and then asks great questions of production and development DBAs. Erik even has a door prize for a sub-set of candidates.
• Deb Melkin thinks from the perspective of the interviewer and interviewee. Talking about disaster stories is always a good team-building exercise (similar to how gallows humor is the greatest form of humor) and it also gives you a feeling for how a person may be able to act in a stressful situation. Also, any DBA who has never accidentally caused a production outage is a time bomb waiting to happen.
• Hugo Kornelis has ascended beyond your mortal plane and has no need for the interview. I did enjoy Hugo going through the thought experiment even with no desire to interview for a job or hire people.
• Brian Kelley wants you to know the soft skills. I do agree that, most of the time, technical knowledge is something you can learn if you’re willing to put in the time and effort. Your starting point can absolutely matter—it’s unlikely I get hired to be a Rust developer on embedded hardware, given the fact that I don’t know the language or much of anything about working in the embedded space—but desire and ability to ask the right questions can go a very long way in separating candidates at the margin.
• Steve Jones has a question for each side of the table. I like both of Steve’s questions a lot.
• Justin Bird has a variant of a Fermi decomposition question. A classic Fermi decomposition question starts with something you couldn’t reasonably expect to know the answer to: How many piano tuners are there in the city of Chicago? But the point of the exercise is to lay out reasonable estimations of various factors that might go into it: your population estimate for Chicago, the average number of people who live in a home (or apartment/condo/whatever), the average percentage of households with pianos, the average amount of time it might take a piano tuner to tune a piano, the number of pianos a piano tuner could reasonably tune in a day, and the number of days a tuner could work in a year. Then you put all of that together to come up with a rough estimate of the number of tuners you’d need in equilibrium, and the rough estimate is probably not that far off from reality. Here’s my dilemma: I don’t like that style of question, myself, certainly not out of left field. But at the same time, Fermi decomposition is a really valuable tool for leads, management, and product managers. So I suppose my compromise is that an interviewer should, like Enrico Fermi himself did, gently walk the person through the process after the initial “I don’t know and have no idea why you would ask me this question” response that I consider eminently reasonable but the average mid-wit interviewer notes as a negative while making a very quiet clucking noise and shaking his head just enough for me to notice before moving on to the next question on his list.
• Chad Callihan puts you in a situation and wants to see how you respond. A tabletop simulation of a situation. Which would be an interesting type of question to game out in a longer interview: something has gone sideways, so what do we do?
• Louis Davidson wants to know what happens when you’re wrong. This is a great signal indicator for candidates and Louis drills into how you can turn a wrong answer into a learning experience, hopefully for the candidate but if not, then for you.
• Deepthi Goguri shares her recent job interview experience and talks about community and the personal side of things.
• Jeff Mlakar sneaks in just under the wire and doesn’t have to make up excuses about how his dog ate his alarm clock and that’s why he was late. Jeff also includes plenty of bad job interview questions that I enjoyed.

Conclusion

Thank you again to everybody who participated in this month’s T-SQL Tuesday.

And the ultimate answer to the question that I’m sure has been burning in your heart for two weeks (assuming you’re going through this live and not viewing the post months or years later)? Fairfield, Iowa.