There are several distinctions in the pool of data scientists available for their expertise which separate them one another. In broad terms, there are two kinds of data scientists out there- vertical data scientists and horizontal data scientists. A vertical data scientist relies on their specialization, which is having very deep knowledge in some narrow field. These comprise of individuals like computer scientists who are familiar with computational complexity of the various sorting algorithms, or statisticians with deep financial insight, software engineers who are accustomed to creating working bodies of code applied to API development and web crawling technology, or even someone with a strong database background for data warehousing and graph databases, and knowledge of Hadoop and expertise in it, even predictive modelling experts who rely on networks, SAS and SVM. Horizontal data scientists are a blend of business analysts, statisticians, computer and software engineers, and domain experts. Their expertise does not necessary lie in complex linear modelling, eigenvalue knowledge, but rather in modern, data-driven techniques that are applicable to unstructured, streaming, a big data such as the very simple and applied analytic bridge theorem to build confidence intervals. They can design efficient, dynamic, simple and robust bodies of code alongside the right data science certifications.
Horizontal data scientists distinguish themselves among other data scientist jobs, as they have features unlike the rest of the talent. Here are some features of horizontal data scientists that make them unique-
- They have some degree of familiarity with the six sigma concepts, more than enough to understand that speed and efficiency is more important than perfection in the realm of analytical concepts.
- They can craft a narrative and success indicators from messy and unorganized data sets, including tangible measurement of success.
- Go about identifying the actual problems to be solved, the data sets that are required, the database structures they need, rather than being passive consumers or using third parties lacking the skill to use/collect data.
- They are well versed in guidelines and what mistakes to avoid, more than a strength in theoretical concepts.
- They possess advanced skill in all aspects of Microsoft Office, especially Excel and other tools that assist in visualization.
- They are known to produce practical and useful dashboards or alternate tools to represent the data found in an effective manner.
- They think outside the box and use their diverse skills to effect the organization in many ways. Creating recommendation engines are made with efficient security in place to detect fake reviews.
- They are often classified as innovators who create practical and useful bodies of code and algorithms serving specific functions. Many employers do not look for this trait for obvious reasons of preferring the good soldier to the disruptive creator.
Many regard vertical data scientists as shortcomings of the educational system that doesn’t account for the cross-disciplinary nature of the profession where the university system trains computer scientists, statistician, and MBA graduate and operations research but not all four into one. Companies are not yet used to identifying horizontal data scientists - the true money makers and ROI generators among analytic professionals. Data science certifications look to bridge this gap in the future, with more comprehensive courses.