Katharina Eggensperger
Professorin at Lamarr-Institut
Dortmund, North Rhine-Westphalia, Germany
Summary
🤩
Rockstar🎓
Top SchoolKatharina Eggensperger is a Professor of AutoML and Hyperparameter Optimization based in Dortmund with 12 years of research and engineering experience bridging academia and open-source. She led an early-career research group on AutoML for Science and completed a PhD focused on automated machine learning workflows, combining deep theoretical knowledge with practical tooling. Her engineering contributions to prominent AutoML projects like SMAC3 and auto-sklearn include EPM implementation, core refactors for numerical robustness, and integrating gradient-boosting and k-NN models with unit tests. Known for shipping production-ready back-end code as well as reproducible science, she blends rigorous evaluation with pragmatic software design. Colleagues describe her as someone who moves seamlessly between research questions and code-level fixes—often spotting numerical edge cases before they surface in experiments.
12 years of coding experience
Master's degree, Computer Science, Master's degree, Computer Science at The University of FreiburgComputer Games Technology, Computer Games Technology at University of the West of ScotlandBachelor's degree, Computer Science and Media, Bachelor's degree, Computer Science and Media at Stuttgart Media UniversityDoktor (Ph.D.), Doktor (Ph.D.) at Albert-Ludwigs-Universität Freiburg
Github Skills (17)
python10
scikit10
machine-learning10
object-oriented-programming10
gradient-boosting10
scikit-learn10
scipy9
unit-testing9
numpy9
hyperparameter-optimization9
automl9
automated-machine-learning9
k-nearest-neighbours9
optimization8
bayesian8
Programming languages (5)
Github contributions (5)
automl/SMAC3
Sep 2015 - Oct 2022
SMAC3: A Versatile Bayesian Optimization Package for Hyperparameter Optimization
Role in this project:
Back-end Developer
Back-end DeveloperContributions:3 releases, 15 reviews, 386 commits in 7 years 2 months
Contributions summary:Katharina's contributions centered on the implementation of the `epm` module within the `smac3` repository, which is designed for automated machine learning and hyperparameter optimization. Their initial work included the creation of a dummy implementation using Python and the integration of existing but incomplete modules. Further commits involved core refactoring of the base and random EPM classes by moving logger initializations to constructors, and refactoring code to address existing versions of numpy/scipy. The user also appears to have started implementing functionality to transform run history data into a format suitable for training an EPM.
bayesian-optimizationhyperparameter-tuninghyperparameterautomated-machine-learninggaussian-process
automl/auto-sklearn
Dec 2014 - May 2022
Automated Machine Learning with scikit-learn
Role in this project:Data Scientist
Data ScientistContributions:46 reviews, 371 commits, 29 PRs in 7 years 6 months
Contributions summary:Katharina's contributions centered around the implementation of a gradient boosting classification model and the integration of a k-nearest neighbors classifier. They added code to build a model using features from the dataset. The user also introduced a unit test suite for the implemented models.
pythonmeta-learningmetalearningbayesian-optimizationhyperparameter-tuning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.