Ramona Bendias
Founding Engineer - Applied ML at Kumo.AI
Berlin Metropolitan Area Germany
Summary
🤩
Rockstar🎓
Top SchoolRamona Bendias is a Founding Engineer specializing in applied machine learning with nine years of experience building practical ML solutions across startups and research institutions in the Berlin area. Trained as a natural scientist with an MSc in Data Science, she blends domain-driven curiosity with production-focused engineering—shipping models, data pipelines, and frontend/backend features in healthcare and AI products. At Kumo.AI and previously Workist she has driven ML projects from prototype to deployment, and her open-source contributions to PyTorch Geometric include adding the PNA model and work on explainability compatibility, showing a strong grounding in graph neural networks. Her bachelor research applied 3D MRI deep learning to Parkinson’s detection, an effort that honed her rigor around negative results and data limitations. Colleagues find her motivated by sustainable, environmentally conscious innovation and adept at turning theoretical knowledge into impactful code.
8 years of coding experience
1 year of employment as a software developer
A-level, A-level at Theodor-Heuss-Gymnasium AalenKTH Royal Institute of TechnologyMSc, Data Science, MSc, Data Science at Technische Universität BerlinLund University
Github Skills (7)
pytorch10
machine-learning10
pytorch-geometric10
deep-learning10
graph-neural-network10
python10
deeplearning-ai8
Programming languages (3)
Github contributions (5)
pyg-team/pytorch_geometric
Oct 2021 - Jan 2023
Graph Neural Network Library for PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:50 reviews, 36 commits, 14 PRs in 1 year 3 months
Contributions summary:Ramona's commits primarily involve the addition and modification of machine learning models within the PyTorch Geometric library. They added the PNA (Principal Neighbourhood Aggregation) model and integrated it into the library, as well as contributing to examples and tests related to GNNExplainer. Their work extended to the implementation of the BAShapes dataset and made adjustments to model compatibility for explainability frameworks like Captum.
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
RBendias/personalized-medicine-project
Mar 2018 - Feb 2022
Code for the project Personalized Medicine - EDAN70 course.
Contributions:16 pushes, 2 branches in 3 years 11 months
pythonpersonalized-medicinedeep-learningmachine-learningmedicine
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.