A Data Mesh Implementation to Facilitate Data Sharing Across Digital Products for Insights Extraction



An important element to improve intelligent healthcare is to develop the medical value of pharmaceutical products based on evidence. Ongoing efforts in our team already produce insights that can enable decision-making to drive medical strategy. These insights are based on collected data (mostly text) fed into state-of-the-art data science algorithms. Interestingly, the most valuable insights are gained when data derive from diverse sources and beyond a single domain. However, a challenge arises when trying to combine all these data prior to further analyses. The data mesh principles of "data as a product" and "self-serve data platform", among others, are applicable to our present use case. We hence implemented a data mesh leveraging the Amazon Web Services Lake Formation fully managed service. Our approach enables us to share data easily and securely among various products of different domains in our organization. This has been an essential step towards avoiding data silos and emerging collaborations across functions.


Dr. Laura Fernandez Gallardo

Senior Machine Learning Engineer
Bayer AG

Laura did her PhD (University of Canberra, Australia) and her postdoc (Technical University of Berlin) applying machine learning to speech signals, performing predictive modeling of users’ characteristics by their voices. In 2018 she started working as a Senior Data Scientist / Data Engineer at areto consulting GmbH implementing end-to-end solutions leveraging data science and cloud technologies. Laura joined Bayer in 2022 as a Senior Machine Learning Engineer. Her main responsibilities include the automation of machine learning workflows and shaping applications which enable data-driven decision making.

Go To Speaker