Why you write code when you interview with Clustrix

Everyone who interviews for a job with Clustrix eventually finds themselves standing at a whiteboard with a marker and a task, “Write some code that will do, X.”

For the kinds of people we’re looking for here, this is their bread and butter. They’ve already written and deployed a lot of software professionally. Or they were at the head of their class at school; coding away on interesting side-projects while other students were sleeping.

But sometimes candidates resent having to do this exercise. Maybe they feel that, given their extensive experience and credentials, it’s demeaning to be solving problems like a schoolchild. I concede that standing at a whiteboard and hand writing code–while someone watches over your shoulder–isn’t necessarily fun or natural, even for a good programmer. So I would like to explain, briefly, our reasoning.

It is well known in the industry that there is a large gap between those who have familiarity with programming, and those who have actually mastered it.[1][2][3] It is surprising how many candidates, with impressive resumes citing many years of experience, are unable to write code to solve simple problems; candidates who are interviewing for a job writing code.

Computers are unforgiving. There is no ‘-ffill-in-the-details’ option to gcc. Making a distributed relational database is very hard. Developers here at Clustrix are expected to swim through code like a duck in water. At Clustrix the engineering team is where the magic happens. It’s a huge responsibility to take ideas/concepts and actually ship working software so that our customers can run their businesses. No posers need apply.

If you are a self-motivated problem-solver who loves programming, we would like to talk with you about making the greatest database system in the world.

young kid working on computer-resized-600

[1] http://norvig.com/21-days.html

[2] http://www.codinghorror.com/blog/2007/02/why-cant-programmers-program.html

[3] http://www.eis.mdx.ac.uk/research/PhDArea/saeed/paper1.pdf