Katharina Beitz


State-of-the-art location optimization platform based in Berlin is searching for a machine-learning-practitioner for incoming projects.
The engineers team itself is well-known for the expertise in statistical and mathematical approaches. As a freelance consultant you challenge the perspectives with hands-on knowledge of possible outcomes in real life, because theory and practice don't always go hand in hand. 


Strategical thinking: how to approach a problem and how to manage a project 0% to 100%
What is a best practice pipeline for solving a prediction problem:
How to clean and reduce the data without losing important information?
How many statistical learning methods should be used?
Data modifications - log transformations etc.
How to avoid overfitting (especially when training data is very specific)?


Sparring partner, project-based 
Professional industry experience related to revenue prediction
Deep statistical understanding of statistical learning (machine learning) methods.
Successful delivery of solutions to prediction problems with the following attributes:
Classification and regression models with high accuracy
Low n large p problems (few observations, many attributes - 'Curse of Dimensionality)
Highly correlated data 

Did we caught your attention? We would be pleased to receive your application or recommendation!