Nowadays, when I see an ad for a "quant" job opening, chances are it’s a developer position — usually a quantitative developer position. In the last six months, it’s becoming rarer and rarer to see a quant position in the traditional sense, i.e., someone who designs and tests models, as opposed to just implementing models. This is not a surprising development, since in these times of uncertainty and big losses, financial firms want people who can quickly improve what the firms already have, and whose work results in immediate usefulness. As with many other industries, in bad times, basic R&D gets thrown out of the door and ignored.
But this change may be permanent for the quant finance profession. After all, it’s unlikely anyone will come up with a revolutionary model like Black-Scholes anytime soon, and you have armies of academics who can devote their time to testing various hypotheses. Wall Street, more than ever, demands immediate results, and results that can boost the bottom line. Gone are the days when physics Ph.D.s were given fat bonuses just to doodle esoteric formulas on company whiteboards.
What this means to quant job seekers is, programming skills — especially in C++, C# and Java — are more important than ever. My own experience has been that, every job I’ve applied to and/or interviewed for required strong programming skills, mostly in C++. Whether you’re a physics Ph.D. student or currently getting your MFE, beef up your C++ knowledge and do lots of programming exercises! (Or go find something else that can leverage your existing skills.)
Good luck, everyone!