Adrian Wolny
Research Scientist at Bayer
Berlin, Germany
Summary
🤩
RockstarAdrian Wolny is a Research Scientist based in Berlin with 11 years of experience at the intersection of machine learning and bioimage analysis, currently working at Bayer. He combines academic rigor from a PhD in ML and a postdoc at EMBL with hands-on engineering, contributing full-stack improvements to ilastik’s GUI and workflows and developing production-ready 3D U-Net models in PyTorch. His work spans model architecture, loss-function design, data augmentation, and usability-driven tooling, showing a rare blend of research depth and practical software craftsmanship. Notably, he focuses on making advanced volumetric segmentation accessible through both algorithmic innovations (e.g., custom losses) and UX improvements in open-source platforms.
11 years of coding experience
Github Skills (10)
semantic-segmentation10
hdf10
pytorch10
work-flow10
machine-learning10
il10
front-end-development10
qt10
data-augmentation10
python10
Programming languages (7)
Github contributions (5)
wolny/pytorch-3dunet
Sep 2018 - Dec 2022
3D U-Net model for volumetric semantic segmentation written in pytorch
Role in this project:
ML Engineer
ML EngineerContributions:24 releases, 14 reviews, 492 commits in 4 years 3 months
Contributions summary:Adrian contributed to the implementation of a 3D U-Net model for volumetric semantic segmentation. Their work included the development and integration of new loss functions, such as DiceLoss and ContrastiveLoss, as well as refactoring and optimizing existing code. They also focused on data augmentation techniques for increased model robustness and the inclusion of hooks to manage the prediction.
semantic-segmentationunet-pytorchresidual-unetu-netvolumetric-data
ilastik/ilastik
Jan 2018 - Nov 2018
ilastik-shell, applets, and workflows to string them together.
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:41 commits, 12 PRs, 4 pushes in 9 months
Contributions summary:Adrian primarily contributed to the Ilastik project by modifying the GUI and underlying functionalities of the application. The commits reveal changes in the `LayerViewerGui`, `CarvingGui`, and `LabelingGui` components, indicating work on user interface elements and workflow logic. The user also addressed a bug related to 3D widget visibility and made adjustments to color table management, indicating a focus on improving the user experience and functionality. These changes are within the context of a machine learning and image analysis platform.
to-stringpythonstringilastikmachine-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.