How to Become a Data Scientist

June 15th, 2012 | Job Search, Press Releases and Industry News | 3 Comments »


For those would be data scientists, here are the skills you need to be a success!

Check here to read the first part about Big Data!

Data scientists are the skilled professionals who can turn raw, unstructured data into answers about brand sentiment, product strategy, asset utilization, and much more. They provide predictive analysis that can pay off in dividends and earn both themselves and their companies a sizable fortune. Now you may be asking, “What do I need to do to pursue a career in data science?” What follows are key skills to have for any aspiring data scientist.

• A great place to start your training is through one of a number of growing university programs which cater to both college age students and ambitious executives looking to increase their business potency. One such program is available through Syracuse University, where they take great pride in providing interdisciplinary training through their School of Information Studies or “iSchool.” A mixture of business and technology, the program exposes students to a variety of disciplines, methods, and skill sets to prepare them for data science in the outside world. Erin Bartolo, the Data Science Program Manager at the iSchool, enthusiastically highlighted the university’s focus on more than just cutting-edge technology. She says, “we’re not married to any particular technical skill […] so if Hadoop disappears and something else replaces it, our students aren’t out of a job. We train them to be thought leaders.” Through Syracuse and other programs, theory and practical laboratory work are combined with soft skills that pay dividends in any position with the ability to communicate, collaborate, and lead.

• For those already outside of college, a great starting degrees for pursuing data science are in Mathematics, Statistics, and Computer Science/Engineering. Additionally, any other degrees and skills you have may give you a unique edge. A broad knowledge base can give you the skills necessary to view big data analysis in a fresh new way!

• It is essential to master a flexible data-intensive program. Hadoop, Oracle, NoSQL, and countless others allow you to mine, index, and aggregate real-time data in substantial ways. As the predominant leader, Hadoop is definitely a good choice, thanks to its high performance velocity and massive parallelization. As an open source project, its massive user base is continually working to streamline operations and evolve possibilities. While Hadoop is in prevalent use among data scientists, that doesn’t mean you should overlook other programs. Alternates like Oracle and NoSQL offer strong scalability and an intuitive layout which is attracting BI analysts and data scientists alike.

• Mining gems out of zettabytes of data is no easy task. It requires persistence to slog through messy, complicated information. Those with experience building databases are situated for greater success because they comprehend the structures and frameworks involved. If you have only performed data management, the transition will not be as smooth. A strong understanding of SQL Server or Oracle provides you with hands-on, in-depth work that can give you an edge in the industry.

A Final Word

If you have passion for and comprehension of big data, there are countless, diverse opportunities out there awaiting you. Positions can be found throughout industries and a great way to learn about new opportunities is to contact one of our recruiters. With new jobs coming through all the time, our recruiters are poised to offer you the guidance you need as you begin a career in Big Data science.

by James Walsh

3 Responses to “How to Become a Data Scientist”

  1. Rob B. says:

    Great article – and the first place I’ve seen “zettabyte” casually used to describe lots of data. Who knew even just a few years ago that a career like this would be possible?

  2. jameswalsh says:

    The field of data science has made incredible leaps and bounds. The ability to assess those massive collections of unstructured data is mind-blowing. Personally, I can’t wait to see the direction and the momentum these types of solutions will gain in the future.

  3. Rob B. says:

    Agreed. My concern will be to see how conclusions are draw beyond trend analysis. For years unrelated data has been used for “evil” purposes. i.e. insurance companies assigning a greater risk to people with red cars. For many, they simply like the color red. The problem with data is that is can be manipulated to be “beautiful in the eye of the beholder”- glass half empty or half full? We must be cautious to not allow ourselves to not be deceived by data without proper cross examination.