Frederike Rueppel has been interested in logic puzzles since she was young. Today, she uses her analytical skills as a data scientist together with her colleagues in the Machine Learning Solutions team at Daimler TSS. The goal: to direct data streams into the right channels in order to enable using the data content. With the results of her work, she is laying the foundation for innovative products and services that make mobility even more secure and pleasant for Daimler customers. In her interview, the data expert explains why artificial intelligence must continue to be explainable, what she finds particularly fascinating about her job, and how data can be used to create real value added.
Ms. Rueppel, as a Data Scientist at Daimler TSS, you develop concepts in order to make large volumes of data useful. What do you like about your work?
I love tackling problems that seem impossible at first. An automotive group like Daimler generates data from a wide range of sources – from planning and production through to our service partners and the vehicles themselves. At Daimler TSS, we help our colleagues from the various specialist units to use this data and information. We combine sources of data, spot connections, and in so doing learn deeper lessons. Our projects frequently lead to entirely new ideas for products and services.
What would be an example of a new service idea?
We are currently dealing with a project for our new electric vehicles. We are developing an application, which projects the capacity utilization of electric charging stations, displays the best charging times for drivers, and guides them to the nearest available charging point.
How do you approach that kind of project?
For example, we use the capacity utilization history provided to us by the charging station operators. We then look at everything, prepare the data sets, and develop projection models. We start very small – for example by taking the charging point capacity utilization for the previous week as a projection for the next week. This is our reference model, and every other, more complex model needs to be better in order to offer added value.
Our work often leads to entirely new ideas for services. We are currently working on an app that projects the capacity utilization of electric charging stations, for example.
What is the next step?
We then refine the concepts and incorporate any current traffic data, for example. The automation begins when we have found a model that achieves the desired level of reliability. The aim is for our app to be constantly supplied with current data. And, as the last step, we integrate the results – for example into vehicles' navigation systems in order to make them visible for the driver.
That sounds very interesting. What has been your favorite project to date?
The project I am currently working on is always my favorite project (laughs). I take a full-on approach to new challenges. Every data source we deal with is different. And when we start a project, the outcome is usually uncertain. At first, we often can't tell what challenges we are going to face. It goes without saying that we want to find the best solution. So it's always interesting, and I like that.
What characterizes Daimler TSS as an employer for you?
I appreciate being treated as an equal. There are no barriers between employees and line managers. I find the open culture important for my work. And as a global group, Daimler offers me the opportunity to advance in my field. One of my personal goals is to support projects even more comprehensively in the future, so that the added value created by data science is more accessible and useful for the specialist units. And my team at Daimler TSS is the right place for that.
Each data collection is different. And when we start a new project, the outcome is usually uncertain. So things are always exciting in my job.
Who is on your team?
I work with 15 colleagues in the Machine Learning Solutions team within the Customer & Car Analytics unit. We are all data scientists, but each one of us has a different background – from data engineering and statistics to IT infrastructure, economics and neuroscience. What makes our team work so well is that we help each other when a certain task requires particular expertise. With major projects, we have been known to work in groups of ten, and we often work with other teams at Daimler TSS such as the Cloud Analytics and Data Integration team.
What is your work day like?
I spend most of my time programming. I particularly enjoy the coordination meetings with my immediate co-workers in my team. We often sit in front of a whiteboard, sketch ideas and develop the project. I always find talking to other people with very different backgrounds and perspectives very fascinating. Another important part of my work is to constantly keep myself up-to-date, from both a subject-related and technical perspective.
And how do you manage to always keep up-to-date?
We follow the current research and discussions in the field in order to be in the know when it comes to artificial intelligence (AI). There regularly are new discoveries and updates, for example with machine learning algorithms and their implementation. That makes data science very interesting and varied. One subject that I am currently dealing with is explainable AI. This is about the requirement not only to develop models for artificial intelligence, but also to make them comprehensible to users. This transparency is important in order to engender lasting trust in AI and its achievements.
Through my work, I want to advance the field of data science. And my team at Daimler TSS is the right place for that.
And on a personal note: Where would you go if time travel were possible?
I would like to go on two journeys through time (laughs). My first destination would be 150 to 200 years in the past, in the time before the industrial revolution. I would like to see and experience for myself how people lived back then. My second journey would take me 150 to 200 years into the future, in order to find out how our technology has developed and how our society has changed.
In person: Frederike Rueppel, Ph.D. (35) Ms. Rueppel has always been fascinated by the possibilities presented by the combination of data and algorithms. That is why the native of North Hesse decided to study mathematics at the Julius Maximilian University of Würzburg, before going on to gain a doctorate there at the Institute for Mathematics. Before joining Daimler, she worked at a consulting start-up helping global corporations to make the right decisions with the help of data. What she particularly appreciates now at Daimler is that the specialist units not only appreciate her as an expert, but also as a co-worker. When she is not extracting added value from large volumes of data in the Machine Learning Solutions team at Daimler TSS, she enjoys playing her violin in amateur orchestras or practicing sports.