This article excerpt is from eFinancialCareers.
Traders and other financial services professionals who know how to code have a leg up on less tech-savvy candidates, as there are an increasing number of jobs available to them as banks and asset managers cut back in other areas. For example, many quant hedge funds have continued their hiring push with an emphasis on candidates who have strong computer skills and investment knowledge. No financial services firm can ignore the need to invest in people that understand technology to keep it at the forefront of the industry and ahead of its competitors.
So you know your way around a computer and you’ve heard of C#, C++, Java, SQL, Hadoop, HTML5, Python and R. But what about Julia? While it’s still early days and Julia is not yet among the most in-demand programming languages on Wall Street, its user base is already approaching 200,000 and its growth rate is doubling every nine months, with hedge funds among the early-adopters.
That means Julia should be on the radar of everyone from traders and operations executives to IT managers, developers and data scientists and really anyone who wants to expand their job options as electronic trading takes over and the industry as a whole becomes more technology-centric.
We spoke with Viral Shah, co-founder of Julia Computing, about the evolution and current status of this fast-growing open-source programming language.
How Did Your Collaboration with MIT Lead to the Creation of Julia Computing?
“We started the project that became the Julia programming language in 2009.
“I am a computational scientist by training. My thesis at the University of California, Santa Barbara focused on parallel computing and it became part of the Star-P product at Interactive Supercomputing Corp. When Microsoft acquired the firm, [co-founder] Jeff Bezanson and I started talking about a new approach to parallel computing. At the same time, [co-founder] Stefan Karpinski and I were discussing the same issues in our research collaboration at UCSB. Jeff asked [co-founder] Alan Edelman, who was also on my thesis committee, about joining the Ph.D. program at MIT, and the four of us came together.
“We set out to solve the ‘two language problem’ back in 2009. Much of our progress in parallel computing was thwarted by the fact that while the users are programming in a high-level language such as R and Python, the performance-critical parts have to be rewritten in C/C++ for performance. This is hugely inefficient, because it introduces human error and wasted effort, slows time to market and allows competitors to leapfrog ahead. This two language problem hinders not just researchers, but also quants, scientists, data scientists and engineers in the industry.
“Many of the earliest adopters were quantitative analysts in the finance industry.”
For more on programming with Julia, including its impact on practical applications and financial-services firms, check out the eFinancialCareers article. It was originally written by Dan Butcher.