En bild på Siddy Persson

29 October, 2019

Data Scientist and Data Engineer - how do the roles differ?

Data, artificial intelligence and machine learning are something that is top of mind for many companies today. But what skills are needed to get started and actually work on it? We talked to our Data Scientist Siddy to learn more.

Hi Siddy, tell me more about yourself!

I would describe myself as a tech nerd with a great passion for technical challenges and engineering. Like to snow in on new technical areas and immerse myself. Last time I looked at how to make almost perfect pizza in a conventional oven. Today I work as a Data Scientist at Digitalent and ended up there because I always liked to solve problems with technology and data. A couple of years ago, I opened my eyes to machine learning and data science and have since strived to put data into work to solve problems and contribute with business benefits.

Data Scientist and Data Engineer - how do the roles differ?

I would probably say that a Data Scientist is expected to have more knowledge in mathematics and statistics. This is probably made especially clear when, for example, is expected to be able to explain why part of a dataset looks in a certain way. It can also be about how to handle data with tricky distributions or largely missing / empty values. A Data engineer is an expert in servicing data and acts as an important facilitator of Data science. This is of course based on the size of the company and the analysis function and outside the industry it is not uncommon for you to be somewhere in between the roles.

Do you need both a Data scientist and Data engineers?

Let's describe it with an example. For example, consider if we want to enable some kind of preaching based on high-volume streaming data and the risk of skewed distributions or missing values. In such a case, a good and close collaboration between Data engineer and Data scientist is required. Furthermore, I would say that one must inevitably be forced to compromise, either on predictive performance or eg. well-built data pipelines, if one chooses to let an individual shoulder both roles.

How can you work together and learn from each other?

Those who have moved more towards the Data Science team learn a lot when working with Data engineers. Both in dialogue and implementations, you open your eyes to powerful frameworks, techniques and best practices. In the same way, I have contributed a lot in discussions that revolve more around how e.g. data looks or how we construct new parameters (features) from existing data.

How do you work within Data with skills development today?

During the autumn, a new competency sharing initiative was launched, something we call Competence Crowds. Here we gather everyone in the company who works in the Data area today, or who want to learn more, and share expertise between each other. So far, we have come up with a brief introduction to both Data engineering and Data science. In the future, we plan to participate in a Data science competition where we solve a real business problem from the ground up.

How do you think the Data area will look like Digitalent in the future?

We will certainly have many subjects to study within the Data crowd. Both Data engineering and Data science are moving forward at a rapid pace and new challenges and frameworks are emerging all the time. I would like to see that we spend a lot of time on understanding to sharpen our analytical capabilities even more rather than necessarily always using the latest framework.

More from Digitalent

Load more