Pascal Iversen
Promovierende R Forscher In at Hasso Plattner Institute
Berlin, Germany
Summary
🤩
Rockstar🎓
Top SchoolPascal Iversen is a doctoral researcher in Berlin with five years of experience applying machine learning and network-based data analysis to precision medicine. With a strong academic record (MSc Data Science, 1.0, and a Biophysics background including ETH Zürich), he bridges rigorous quantitative modeling and practical forecasting. He contributed to AWS’s GluonTS open-source library, improving probabilistic time-series distributions (including a NanMixture to handle missing data) during an applied science internship, showing attention to edge cases and statistical correctness. Pascal’s work spans academia and industry—from BASF and AWS to Hasso Plattner Institute—combining deep probabilistic forecasting, teaching, and reproducible research. Colleagues describe him as a careful implementer who turns theoretical models into robust, test-covered code for real-world biomedical problems.
5 years of coding experience
1 year of employment as a software developer
Master of Science - MS, Data Science, 1.0, with Distinction, Master of Science - MS, Data Science, 1.0, with Distinction at Universität PotsdamBachelor of Science, Biophysics, 1.2, Bachelor of Science, Biophysics, 1.2 at Humboldt-Universität zu BerlinExchange semester, Biophysics, Exchange semester, Biophysics at ETH Zürich
Github Skills (12)
forecasting10
mxnet10
machine-learning10
deeplearning-ai10
forecast10
time-series10
deep-learning10
statistical-models10
python10
data-science10
artificial-neural-networks9
neural-network9
Programming languages (5)
Github contributions (5)
awslabs/gluonts
Jun 2020 - May 2021
Probabilistic time series modeling in Python
Role in this project:
Data Scientist
Data ScientistContributions:19 reviews, 16 commits, 22 PRs in 10 months
Contributions summary:Pascal contributed to the development and testing of probabilistic time series models within the GluonTS repository. Their work focused on correcting and improving the statistical calculations of distributions, specifically the MixtureDistribution. They implemented and tested the NanMixture distribution, designed to handle missing values, and addressed various edge cases in log probability computations. Furthermore, the user fixed a typing error in a DeepAR file and added functionality for a beta distribution.
forecastingpythontime-series-analysistimeseries-forecastingaws
PascalIversen/gluon-ts
Jun 2020 - Dec 2020
Probabilistic time series modeling in Python
Contributions:130 pushes, 36 branches in 5 months
dtwpythontime-series-analysisdata-sciencemachine-learning
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.