The Data Scientist profile is a quite new and “a bit” complex combination of qualities, as you can see in the last post  “Anatomy of the Data Scientist”. Given the difficulty of finding experts in all the above-mentioned skills, a series of ad hoc profiles are created in a natural way that aim to cover different needs. There does not seem to be a consensus in the technological community about the number of profiles existing within the data analytics but rather that these profiles are designed on the field according to the requirements of the project.

Profile examples according to experts

To realize the importance of designing a concrete profiling according to needs, we took the following example exposed in the Big Data Spain2019 event, where Elena Grewal, Head of Data Science on Airbnb, described three types of roles within their teams: the Data Analytics, which defines and monitors the metrics, creating narrative with the data and controlling the treatment tools of these; the Data Algorithms, which develops the algorithms to improve its products; and the Data Inference, which establishes causal relationships with statistics.

In this other case, Tauhida Parveen, leader of the educational startup Thinkful, in the Big Data Florida event of 2018, indicates that there are four profiles of Data Scientist:

• The super-analyst: Experts in problem solving, good communicators and product builders.
• The statistics-man: Analyzes statistical and probability mathematical models.
• The guru:Expert in programming machine learning algorithms for automation processes.
• The researcher:They look for what will be the next step in the Big Data world, a profile closer to innovation.

There are many proposals, but if we collect the functions that are most often repeated in different business contexts, these could be reduced to the following three profiles:

• Mathematician and statistician: Has sufficient knowledge to develop mathematical and probability models and knows different types of algorithms to use in accordance to the AI, Machine Learning or Deep Learning problem.
• Data Engineer: He has computational thinking and knows the tools that allow him to solve different problems in the massive treatment of different nature data.
• Decision scientists: They are usually responsible for making decisions. They know how to interpret the data altogether with the business context, they have communication and management skills, with recommended sociological knowledge, they know how to create narrative with data and visualize it in order to achieve a better understanding.

Every profile’s skills are commonly shared and could overlap (or should) with those of other profiles.

Required collaboration

With the expectation that companies have of finding complete profiles of Data Scientist, the reality we find in the market is very different, where recruitment and retention of talent is becoming a real challenge for digital companies.

This is why the collaborative perspective seems to be the best solution in the project developments that involve data processing in a context of complexity, and it seems then essential to consider the combination of multidisciplinary efforts together with the flexibility to create balanced teams that can cover the diverse needs required, which determine the success of the project.

To conclude, nothing better than an illustration that exemplifies the enriching of the collaboration between professionals of Data Scientists.