Konstantinos Benidis
Sr. Applied Scientist at Amazon Web Services (AWS)
Berlin, Germany
Summary
🤩
Rockstar🎓
Top SchoolKonstantinos Benidis is a senior applied scientist based in Berlin with 9 years of experience applying deep learning, probabilistic time-series forecasting, and optimization to real-world problems. He has a strong research-to-production track record at AWS and Amazon AI Labs, contributing to forecasting products and Amazon Q while bridging research science and engineering. His PhD in Financial Engineering and background in information theory and communications underpin a rigorous, mathematically grounded approach to model design and algorithmic improvements. An active open-source contributor to the widely used GluonTS library, he has improved core probabilistic distributions and inference reliability—work that directly strengthens production forecasting toolchains. He combines academic depth with hands-on engineering, often focusing on numerical robustness and API usability that are easy to overlook but crucial in production systems.
9 years of coding experience
6 years of employment as a software developer
Master Thesis, Information Theory, Master Thesis, Information Theory at Technische Universität MünchenUPC Universitat Politècnica de CatalunyaDipl.- Ing. Electrical and Computer Engineering, Dipl.- Ing. Electrical and Computer Engineering at National Technical University of AthensHong Kong University of Science and Technology (HKUST)
Greek, English, Spanish
Github Skills (8)
forecasting10
mxnet10
forecast10
time-series-forecasting10
deep-learning10
python10
machine-learning9
pytorch8
Programming languages (4)
Github contributions (5)
awslabs/gluonts
May 2019 - May 2021
Probabilistic time series modeling in Python
Role in this project:
Data Scientist
Data ScientistContributions:36 reviews, 25 commits, 43 PRs in 1 year 11 months
Contributions summary:Konstantinos primarily contributed to the development and improvement of probabilistic time series modeling within the GluonTS repository. Their work involved refactoring and fixing bugs in the `Binned` distribution, addressing issues in inference tests, and making API improvements related to distributions. They also made significant contributions to the `PiecewiseLinear` distribution, fixing bugs and adding tests. These changes indicate an active role in enhancing the core functionality and reliability of the probabilistic time series models.
forecastingpythontime-series-analysistimeseries-forecastingaws
benidis/gluon-ts
Jun 2019 - Feb 2021
GluonTS - Probabilistic Time Series Modeling in Python
Contributions:83 pushes, 46 branches in 1 year 9 months
forecastingpythontime-series-analysismachine-learningpredictions
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.