Data Science Career Paths - An Overview

Data Science

Data science is booming, and there’s a good reason for it. We live in a data rich world and there is a growing need to turn all this data into information, so that we can constantly improve what we’re doing. Education is one of the sectors that can immensely benefit from data science. We have plenty of data about our students in the form of exam scores, their online activities in learning management systems, or LMS, social media, submitted homework assignments et cetera.

If we had adequate resources to help with analyzing the student data in our possession I’m pretty sure that we could quickly do some innovative things in improving the lives of our students and enhancing the chance of their academic success. However, the reality is that we are not able to keep up with the demand for data science professionals and a lot of our visionary projects are put on hold.

This is why we need more professionals in data science. In IT, developers can make a difference by automating various aspects of data science, especially those labor intensive and mundane ones. Among the options available today, machine learning seems to be one of the most promising technologies, and this trend presents a relatively new data science career path in machine learning. Since there are so many software tools that work as enablers in data science, it’s now mandatory for data scientists to almost double as IT developers.

A bare minimum requirement is ability to be able to program in popular data science computer languages such as Python or R. If your job deals with big data, you need to be familiar with technologies such as Hadoop, MapReduce and Spark. Data scientists also play the role of an interpreter between highly technical domain experts and lay people who consume the results of a data analysis effort.

To be effective in your interpreter role, you need to be able to use powerful visual aids, which is why there is a data science specialty focusing on visualization. Data science is getting a lot of traction these days and you will have ample opportunity to develop your career in this newly emerging and evolving field.

Careers in Data Science

Data Scientist

The data scientist’s job title is quite broad by design. These days organizations of all sizes need help with turning their data into actionable intelligence. Bigger companies can afford to hire multiple employees to work on their data science projects, each focusing on different aspects such as data cleaning and analysis, machine learning, and visualization. Smaller companies are lucky to have a dedicated data scientist position.

The data scientist job titles comes handy for these smaller scale operations that are only able to hire generalists rather than multiple specialists. Therefore, although lacking depth, a data scientist should be able to deal with almost any aspect of data science tasks mentioned earlier. Let’s say that your boss wants to know how the salaries of his employees compare to those of people working at other departments in comparable positions.

You initially need to wear a data science analyst hat to locate the data you want and prepare it for further analysis. Luckily, there are a number of inexpensive commodity data science services available in the cloud today. These include Amazon Quickstart, IBM Data Science Experience and Microsoft Azure Data Science Virtual Machine.

Armed with these affordable tools at your fingertips, you can proceed with your data analysis. In this scenario, you may not need heavy-duty, big data services at all but can still benefit from reporting and visualization functions available in the cloud. The data science position described in this scenario is an excellent opportunity to show off your IT developer skills. Because it challenges you in many areas of your training in IT development, like programming skills, knowledge in software tools, and on-the-job domain knowledge training.

If you are a recent graduate of a four year applied data science program, you are perfectly suited for this jack of all trades position. If you have a background in IT or computer science, you also have a great potential to be successful in an entry-level data scientist position. You just have to be willing to do a lot of self-study in data analysis and data science tools to build on your IT foundation.

Data Analyst

If you despise the idea of being a generalist, you have ample opportunities to become a specialist in Data Science. One such option is becoming a Data Analyst. As its name suggests, this job focuses on preparing, processing, and analyzing data. To excel in this position, you need to be well-versed in statistics, scripting languages like Python, and data analysis packages such as, Excel, SPSS, SAS, and R.

If you’re dealing with big data and need to do some heavy lifting in distributed data processing, you should at least know how to prepare your data analysis jobs to be worked on popular distributed processing platforms such as MapReduce and Spark. The nice thing about being a data analyst is that you can just concentrate on what you do best. That is, data analysis, and don’t worry about other things that must happen in a typical Data Science workflow.

These include all the IT Plumbing work of database management, installing and maintaining statistics packages, cloud computing, and distributed processing platforms, etc. If you’re ambitious and would like to expand your territory into predictive data analysis using Big Data Analytics, you’re well-positioned to develop your career into a Data Machine Learning Specialist.

This transition may not be so difficult because there are already lots of great tools you can play with to learn the basics. I recommend that you start with Weka, which features powerful machine learning libraries with an intuitive graphical user interface or GUI. Have you made up your mind yet? Do you want to become a generalist or specialist? I’m sure you’ll be fine no matter what you choose to do.

Machine Learning

Machine learning scientist job is a step up from a data analyst position. This is further specialization of an already highly technical position. Because of this advanced nature of the position, it requires decades of training to become an experienced machine learning scientist. Basic machine learning tasks such as regression and clustering demand little manual intervention.

There are already well-known algorithms to get the job done, however not every problem can be solved in an optimal way using the existing consumer-grade machine learning solutions. Here, the key word is optimization. Take the problem of image recognition. We’re already pretty good at it, but the solutions are not perfect and there is still room for improvement. This is why they still have competitions for custom image recognition.

In this example the challenge is to detect all ships in satellite images as quickly as possible. You cannot solve this type of machine learning problem by simply reusing the existing code. You must at least tweak it, or develop an entirely new model and its implementation to improve performance drastically. This is where your expertise as an IT developer shines. Surely, you do still have to have the right kind of highly special skills to pull this off.

General IT developer knowledge wouldn’t cut it, but your training in computer science and IT will serve as a great foundation to build on. The ship identification test is left for the very best of IT developer mind. Therefore, if your job title is machine learning scientist, you are at the top of the IT developer food chain.


A picture is worth a thousand words. When done properly, Visualization can be one of the most effective tools to convey the message your data is sending implicitly. Therefore, the role of a Visualization Specialist, is that of a Communicator or Translator. In this case, the communication, or translation, occurs through visual aids.

I’d like to remind you that, a majority of consumers of a data analysis, are laypeople, including managers, customers, decision makers, and so on. Your job as a Visualization Specialist, is to interpret the Numerical and often Cryptic data analysis results into an Intuitive graphic that truly conveys the most important information relevant to the data consumer needs.

Also necessary are Programming skills, and Knowledge in database management systems, because tools cannot always do everything for you. As you might already have guessed, other important qualities for Visualization Specialists are being artistic and creative. If what I have described so far is close to your aptitude and aspirations, don’t hesitate to start your journey into this exciting Visualization Specialist career today.

We certainly need more people like you in the data science field.

Preparation Tips

As the demand for data science increases, more and more colleges and universities are starting to offer undergraduate degree programs in data science. There are also a number of schools providing graduate programs in data science. There are pros and cons associated with starting early or later when it comes to your pursuit of a data science degree. Getting an early start as a college freshman means that you can develop your skills as a data scientist in a reasonable pace and in a balanced way, because you have four years to study various aspects of data science and get plenty of exposure to all the essential topics, tools, and technologies involved in data science.

This is a great option for those who aspire to be data scientists and data analysts. After getting a job in a particular field, such as business intelligence, or BI, you can develop your domain specific knowledge through your on the job training opportunities. This learning by doing approach is fine and can give you more options in terms of your job search because you can get a job in virtually any industries. Data science is a foundational topic like mathematics.

But Data Science is only meaningful when it’s put into a context to solve problems in a specific domain. Therefore domain knowledge is critical for data science to keep its relevance. If you are a data scientist working in the cybersecurity industry, gaining expertise in networking or firewalls is highly beneficial because it gives you more insights when you’re trying to do your job as a data scientist.

This is why it may be beneficial for you to get your undergraduate degree in a non-data science major like cybersecurity or business and then get your master’s degree in data science. For those of you who are overachievers, a double major is also an option. If you are already a working professional in a particular industry, getting a master’s degree in data science is a way to enhance your career opportunities.

As you can see, there is no right time to jump in to data science no matter where you are in your career development process.


Data science is still evolving and so are it’s certifications.

Vendor neutral certifications are starting to appear, one of them is Certfied Analytics Professional, or CAP. The Analytics Certification Board oversees the CAP program and consists of members from both academia and industry. There are also a number of universities offering data science certificates.

Because of it’s technology centered nature, data science industry has a number of vendor specific certifications. Companies like Microsoft, Cloudera, EMC, Oracle and SAS all have their own data science certification programs. Microsoft has it’s Microsoft Certified Solutions Expert, or MCSE, data management and analytics. Cloudera offers it’s Cloudera Certified Professional, or CCP, data engineer certificate.

Other certification opportunities include

  • EMC Data Science Associate or EMCDSA
  • Oracle, Business Intelligence Certificate
  • SAS Certified Data Scientist

Although not as mature as certification programs in other fields with longer history, data science certifications still provide great value for professionals who are interested in advancing their careers and taking them to the next level.

Write a Comment