Copying standard models? I think anyone can do that. Thanks to my technical creativity, I can easily develop (predictive) models and think of the right variables to go with them.
Whether during a meeting or in given information; I quickly grasp what is going on. Because I recognize certain connections quickly, I can easily get to the heart of a problem.
Solely mastering a programming language is not enough. I can make a good estimation and design beforehand, which allows me to write structured code.
Where this is not self-evident for most data fanatics, I like to connect with others. For example, I like to collaborate and yes, I also like to attend Friday afternoon drinks!
Besides a Data freak, I am…
My big dream is to see a tornado in real-life. Truly! I even dream about tornado season. On the work front, I am interested in having my own business. And well, that brings me back to data. Together with a colleague I recently developed my own product (a trading bot). This tastes like more.
I have worked on this project:
Data Scientist at DSM
At DSM I worked as a Data Scientist. Many valuable DSM-employees decide to quit voluntarily. This is seen as a major problem because on the one hand, a lot of knowledge is lost, and on the other hand, new (good) employees must be found and onboarded. To prevent so many employees from leaving, Vivien Pfannstiel (former Vlammr) and I developed an Extreme Gradiant Boosting model. This model can be used to indicate the risk of employees quitting for every department. These insights are, in turn, used to take targeted measures to prevent valuable employees from quitting.
When people want to learn programming, they often start with a course on YouTube, for example. Nothing wrong with that, but you only get good by practicing a lot! My tip? Try solving the problems on project Euler; projecteuler.net. For me, this has been an amazing exercise!