In my second year of grad school (1987-88), my thesis advisor asked me to develop some PC-based digital signal processing (DSP) tools to be used for computer-based lab assignments. (At the time, PC-based engineering labs were very new, laptops weren't a thing yet, and Student Edition of MATLAB was still a couple of years away from exploding onto the academic scene.) I tackled this project enthusiastically, implementing a couple of dozen operations that dealt with such things as polynomials and roots, complex arithmetic, filtering, quantization, up- and down-sampling, Fourier transforms, and plotting. These operations could be combined in simple scripts to solve homework problems. My pleasure and pride in this work is demonstrated by the fact that I still, after many years, have the source code, user's guide, and programmer's guide in my file cabinet scanned in the cloud. (My wife wanted to know how this project was getting me closer to my Ph.D., which I have to admit was a good question that didn't have a good answer.)

I think about this experience whenever I interview a candidate for a software development job. The MathWorks has many products in specialized areas of engineering, science, and mathematics, so we often hire domain experts in these areas to develop software. Our biggest challenge in evaluating a candidate is to answer this critical question:

Do they have the tool builder's gene?

Here's what I mean: Many candidates can describe something that they have created that was useful to someone else. But only a few find the tool-building activity itself to be intensely rewarding. Only a few enjoy extending the tool so that it's useful not only to themselves and maybe a few others in the lab, but to a wide audience. Only a few are motivated to refine and extend a tool after it has solved the problem for which it was originally created. Those with the tool builder's gene get their energy from knowing they've helped many people do their own jobs better.

From my sophomore year in college through getting my Ph.D., I focused single-mindedly on becoming an engineering professor, and I did become one for a few years. But I was fortunate to realize quickly that academia ultimately wasn't the best place for me, and that I would be much happier at a place like The MathWorks. The reason is that I have the gene.

How about you? Are you interested in working at MathWorks? Then please visit our careers page. And here's a hint: the Engineering Development Group is a great way to get started on a technical career at MathWorks.

PS. Hey, the blog comment section has a new look and features, including better formatting, threaded replies, and notifications. Try it out: Share with us a story about your favorite tool you built for others to use.