Stefan Brunhuber

I am Stefan, working as a Lead Data Scientist & Engineer to tackle Climate Change. I am an experienced analytical, and technical problem solver with a background in Atmospheric Sciences. To find out more about me and my projects visit my personal website and blog www.stefanbrunhuber.com I am particulary interested in things that matter like e.g. Climate Change, New Technology, and Digitalization. Tackling any of these issues by means of Data Science, Machine Learning, Statistical Modeling, Data Visualization, and Software Engineering mixed with Creativity and Curiousity I find even more exciting. Topics I worked on: ● Quantifying Climate Change Risk ● Data Science Apps with RShiny - APIs + Data-Pipelines + Databases + Analytics + ML-Models + Visualizaton + Deployment (Docker, Gitlab) ● Product Recommendation Engines (Conception & Model Training & Deployment) - Collaborative Filtering ● Coordination & Execution of Data Science & ML projects with external partners (Universities, etc) Models I used - Classification Models (Decision Trees, SVM, CatBoost/Shap) - Clustering Models (PCA, K-Means, DBSCAN) - Collaborative Filtering - Deep Learning w. Keras (Sequence Clustering - LSTM & CNN) - Time Series Analysis and Forecasting (Bayesian, GAM, ARIMA, AutoCorrelation) - Extreme value theory (GEVD, POT) - Generalized Additive Models (GAM) ● Current Tech Stack - Linux / Bash - Python: numpy, pandas, jupyter/jupyterLab, sklearn, scipy, matplotlib, fbprophet, google-api-python-client, implicit, lightFM, Keras, CatBoost, ... - R: Shiny, tidyverse (ggplot2, dplyr), randomForest, kmeans, googleAnalyticsR, causalImpact, prophet, reticulate, rgdal, leaflet, ... - Gitlab/Git - Docker - Flask - MySQL