I enjoy developing Open Source software and currently do so mostly in R and Python.
During my PhD, I have (co-)developed several extensions for the mlr3 ecosystem and contributed to several others.

mlr3

• mlr3pipelines: Machine learning pipelines for mlr3.
  M. Binder, F. Pfisterer, M. Lang, L. Schneider, L. Kotthoff, and B. Bischl. mlr3pipelines - flexible machine learning pipelines in R. Journal of Machine Learning Research, 22(184):1–7, 2021

• mlr3fairness: Bias Audits for mlr3.
  F. Pfisterer, S. Wei, S. Vollmer M. Lang and B. Bischl. Fairness Audits And Debiasing Using mlr3fairness

• mlr3keras: Deep Learning with keras and mlr3.
  A first attempt to integrate keras with mlr3. Superseded by mlr3torch.

• mlr3torch: Deep Learning with torch and mlr3.
  I help developing mlr3torch mainly in a supervisory capacity.

R

• mlr3pipelines Machine learning pipelines for mlr3.
• mlr3fairness Auditing, visualization and bias mitigation for learners from mlr3.
• Deepregression: Fitting Semi-Structured Deep Distributional Regression in R
• mcboost: Multi-calibration and multi-accuracy boosting.
  F. Pfisterer, C. Kern, S. Dandl, M. Sun, M. P. Kim, and B. Bischl. mcboost: Multi- calibration boosting for R. Journal of Open Source Software, 6(64):3453, 2021
• AutoxgboostMC: A simplistic AutoML system based on tuning xgboost models.
• YAHPO Gym: An R Interface to the Pyhon YAHPO Gym library.

Python

• YAHPO Gym: A multi-objective multi-fidelity benchmark for HPO optimizers. For this paper, I developed software to interface surrogates and a tiny AutoML system that allows fitting surrogate models for use with YAHPO Gym called yahpo_train.

Web

• MCML: I created the website for the Munich Center of Machine Learing (MCML).