Comment by peacemaker | Hacker News Reader

Comment by peacemaker | Hacker News Reader

As a senior guy who's been programming a long time, I honestly don't know what I'd pick in answer to this question. Partly because I've done loads of cool (and no so cool) things over the years and partly because I don't have the best memory in the world.

How do you decide which past accomplishment is worthy of a 30 minute conversation? Usually what comes to mind when asked this question are simply the most recent things I've worked on. If you've had the misfortune of doing crappy work for the last few months then it makes it a bit harder.

I've seen lots of these kinds of posts, trying to "solve" interviewing. It's a topic close to me right now as I'm in the middle of the interview cycle myself. I just don't think it can be solved. There are too many variables and too many non-scientific aspects to it. Most of the time interviews are decided in that first impression with your interviewer - it feels like a primal instinct people have about other people.

One thing I can say though, is that you aren't going to find the perfect candidate with just one question.

There are a whole bunch of "Perfect Interview Questions" that merely filter out people who have never heard them before. This is the latest on the list.

I'm in job-hunt mode so I guess I'm going to sit down and prepare an answer for this one, too.

Yep. This question is actually:

"Tell me about a project in which you demonstrate the skills and character attributes that I am looking for without me telling you what those are."

A good interviewer will ask questions like this only having given you enough context to choose the most appropriate example.

A bad interviewer will not, instead believing that the less information they give you the more revealing your answers will be. All they achieve is to introduce more chance into the interview process by leading you to pick less relevant examples.

I've been a pro for nearly 15 years and the project that I'm most proud of was something i did as an undergrad. Now I'm sad.

Yeah, if I had to answer "what project did you work on that most changed the life of your customers" it would be my first post-college job where I created a market space that didn't exist before.

But, like you, what does that say about the next 15 years of my career? I'm a waaaaay better developer than I was at the end of my first job.

So say exactly that. It sounds like a great answer to me.

Exactly.

It is important we understand the words that we left out of the original article.

The point of the question isn't to get an impressive project. While we, the interviewee would love to say, guidance computer on Voyager or the Ray Tracer used in Toy Story; the point of the question is get to know the candidate and how they operate.

Yup, me too. And it's the worst code I've ever written. Miranda IM. Millions of downloads, popular for 14 years. I got bigger than I ever imagined it could.

What have I worked on since: ACDZip, ACDSee, Tracktion, EAW Resolution, SPIN Review. ACDSee was a brief success, but the rest - meh.

At least your failed projects have names, mine never even got that far! But seriously, how should you answer the question if the honest answer is something you did 10+ years ago? I could talk about that project all day long, but I can sense a "What have you done for me lately" reaction if all I have to hang my hat on is something that happened when Bill Clinton was President!

Knowing that would make me more likely to hire you. On the one hand, you recognize it was the worst code, and you've improved since. On the other hand, you recognize that going into an ivory tower and designing perfect code, is not necessarily required for success.

Any given project will be calibrated somewhere between the extremes of "just ship it" and "take 5 years to make a perfect 1.0". A developer who, based on experience, is flexible enough to calibrate themselves to match the project -- who can adapt to reasonable balances of "do it faster" vs. "do it better" -- is valuable.

Thank you for sharing this, peacemaker and FigBug.

This magic-bullet interview question had me depressed and made me feel like a failure. I too have nearly 15 years of professional experience. My last projects that I can show off were 4 and 8 years ago.

I don't feel like a failure anymore. I got a reminder that success depends on blind luck a lot and is rather overrated. Hard work, improvement and patience are still essential.

P.S. I have used both Miranda and ACDSee in the past. They reached my highest criteria for software, that few applications reach – were small, fast, reliable and very useful. I wish FigBug best of luck.

Yeah its very situation dependent. If I get an interview at the small motor sports place near milton keynes (its not Aston Martin -:) )

I would probably use some of the stuff I did on back at my first job working in the mathematical modeling section of a world ranked RnD organization.

If it was big data I would use my experience back in the 80's doing map reduce for British telecom - and that proves i can ship :-)

Reply on news.ycombinator.com