I am currently based in Berlin (Germany) where I work as a Data Engineer at Haensel AMS. Initially, I worked on Data Science projects and used predictive and data-driven approaches to solve problems for clients in different industries, such as e-commerce, logistics, and retail. However, my focus has now shifted to the data engineering side of things. I primarily build ETL/ELT pipelines that integrate various data sources, and I enjoy the challenge of finding the optimal solution for each task while considering the problem's limitations and context.
Working as data engineer on ETL/ELT pipelines that transform and join various data sources for (marketing) reporting purposes. Depending on the requirements, this involves accessing data on AWS/GCP/Azure, transforming using Python or SQL and orchestration via Airflow.
Working across various industries (e-commerce, logistics, retail) on data-driven solutions. My role involves communicating with clients to work out solutions to their problems, subsequently developing and deploying them in a cloud environment.
Master track that teaches service design, business intelligence and business analytics with the
overarching goal of creating valuable, data-driven and smart services for organizations.
The studied subjects are applied in practical projects for companies located in the Limburg area. See
More!
Bachelor programme that tackles the strategy, resources and management required to run a business.
Furthermore, it teaches the professional skills to put business ideas into practice, including teamwork,
conflict resolution, leadership and presentation skills. See
More!.
In cooperation with the BISS Institute and Open University (NL), I wrote my Master Thesis on the applicability of using Deep Learning methods (Long Short Term Memory networks to be specific) to predict a students learning trajectory. For this task I pursued an extensive descriptive analysis of student behavioural data (e.g. course durations/grades) and ultimately programmed Machine Learning models to predict for a given student his/her future propensity to drop-out of the programme. During this process I gained invaluable practical experience in handling and manipulating large datasets (performed using Python and Pandas) and also consolidated my expertise in training and evaluating Machine Learning algorithms.
In the world of pricing, there are many different approaches used to effectively set prices in the given market context. In some cases, there is a limited amount of goods/services available that need to be utilized at a fixed point in time (e.g. airline tickets or hotel rooms). While in other cases, there are no such capacity constraints and the goal might be to maximize overall profit by selling as much of ones goods as possible. At Haensel AMS, we are hosting the Dynamic Pricing Competition to create a benchmarking tool for practitioners and researchers in the field of pricing to gauge the performance of their pricing algorithms. At the Applied Machine Learning Days 2020 in Lausanne, I gave a presentation on the set-up and key findings of the preceeding Dynamic Pricing Competition.