Bart Kastermans' Home Page

Bart Kastermans is a software engineer and data scientist based in the Netherlands (working in Amsterdam area, living in Uitgeest). In 2012 he made the switch from an academic position (Mathematics) to a position in IT. Since then he has developed into the tech lead of the team that is building a streaming data engine based on kafka and flink running on kubernetes at KPN. Next to building the team, and with that team selecting/designing the architecture, it means doing the engineering. As some examples this invovled writing flink jobs (scala), working on logging and monitoring, configuring components to use SSL, making design choices to ensure manageability of the cluster, writing parts of the custom build components (java), generic cluster management (bash, python), ensuring customer privacy is respected and the privacy policies are followed.

He is currently doing the Udacity nanodegree on Deep Reinforcement Learning. Reading the book Introduction to Cryptography by Katz and Lindell. And learning to use wireshark better to check that the security settings in the streaming engine do indeed have the desired effects (of encrypting the inter pod traffic). Also he is learning to sail in his jeanneau sun 2000 sailboat.

Having been at the current position for about three years he is starting to look for new opportunities. Next to data science and data engineering he is very interested in security and cryptography; which he would enjoy to be part of a new challenge. Team leading or management are neither his strong points, nor his interest. It is a technical position he is interested in.

At the University of Michigan, Ann Arbor, he wrote a thesis in Set Theory titled Cofinitary Groups and Other Almost Disjoint Families under supervision of Andreas Blass and Yi Zhang. During the writing he visited Yi Zhang at Sun-Yat Sen University in China for a year.

After getting a PhD he worked as a postdoc at the Mathmatics department of the University of Wisconsin, Madison, for three years. During this time he worked with Steffen Lempp, resulting in a couple of papers among which is the paper on Comparing Notions of Randomness. He then moved on to be an Assistant Professor at the Mathematics department of the University of Colorado, Boulder, for three years. This was a tenure-track position, but after those three years he decided to give in to a long standing interest in information technology.

His email is