Making the most of ML for real-world data

I enjoy exploring how models interact with the data in order to make them useful in practice.

To build more useful models, I believe a combination of different parts is needed:

• Understanding of the data and the problem via exchange with domain experts and proactive data exploration.
• Knowledge about how models work and critical thinking about what may go wrong.
• Good metrics and benchmarks that measure what we actually care about.
• Easily (re)usable code base and data.

My favourite ML quote: “All models are wrong, but some are useful”

Other interests

I am interested in entrepreneurship. However, I decided to first probe the industry world by joining a well-established company.

And of course: I enjoy participating in scientific communication. - As you may have figured out from this blog. While my interest is broader than this blog, there is sadly never enough time to do everything.

CV

For my up-to-date CV please refer to LinkedIn.