Johannes Pawelczyk
Digital Solutions Manager at Ortivity
Munich, Bavaria, Germany
Summary
🤩
Rockstar🎓
Top SchoolJohannes Pawelczyk is a medical doctor turned digital solutions manager based in Munich who bridges orthopedics, operational excellence, and applied AI to streamline clinical workflows across 120+ outpatient centers. With 11 years of experience, he supports strategic digitalization at Ortivity while leading research efforts at OrthoCoPilot to translate ML and automation into clinical impact. He brings hands-on technical chops from open-source work on PyTorch Geometric—optimizing Graph Diffusion Convolution performance and parallelizing approximate PPR—to inform scalable ML-for-graphs solutions in healthcare. His background as a surgical assistant, international fellow, and research fellow gives him rare credibility in both operating rooms and engineering discussions. Colleagues rely on him to connect clinicians, IT, and leadership to deploy usable tech rather than just prototypes. He combines rigorous academic training (Heidelberg, TUM) with a pragmatic, patient-centered drive to make digital health tools truly operational.
11 years of coding experience
Doctor of Medicine - MD Medicine, Doctor of Medicine - MD Medicine at Heidelberg UniversityDr. med. Medicine, Dr. med. Medicine at Technical University of Munich
Github Skills (19)
pytorch10
python10
deep-learning10
graph-neural-network10
graph-convolutional-networks10
data-structure9
operation9
scipy9
algorithm9
algorithms9
ml9
numpy9
tensorrt9
tensorflow9
mle9
Programming languages (12)
Github contributions (5)
pyg-team/pytorch_geometric
Nov 2019 - Aug 2020
Graph Neural Network Library for PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:17 commits, 5 PRs, 7 comments in 9 months
Contributions summary:Johannes primarily focused on implementing and improving the Graph Diffusion Convolution (GDC) transform within the PyTorch Geometric library. Their contributions included adding the GDC functionality, addressing performance issues by parallelizing approximate PPR calculation, and refining the implementation by incorporating features like optional normalization and specifying top-k axis for sparsification. They also made modifications to existing graph utility functions and examples related to GDC, reflecting their involvement in integrating and optimizing this graph preprocessing technique.
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
gasteigerjo/gtn
Jul 2021 - Apr 2022
Contributions:8 commits, 5 pushes, 1 branch in 8 months
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.