Florian Scheidl
Research Scientist at Forschungszentrum Jülich
Aachen, North Rhine-Westphalia, Germany
Summary
🤩
Rockstar🎓
Top SchoolFlorian Scheidl is a research scientist with five years of experience building and scaling AI-driven weather and energy forecasting systems. He combines a strong mathematics and data science background from ETH Zürich with hands-on MLOps, feature engineering, and high-performance computing work—most recently optimizing AI weather models at Forschungszentrum Jülich. His prior research on neural compression for weather reanalysis and practical experience in power forecasting and smart panel savings modeling reflect a focus on making large-scale geoscientific data both efficient and actionable. Based in Aachen, he bridges academic rigor and production engineering, often applying graph neural networks and compression techniques to real-world environmental problems. Notably, his career path shows a recurring emphasis on squeezing performance and cost out of large scientific datasets while keeping models deployable.
5 years of coding experience
3 years of employment as a software developer
Master of Science - MSc, Data Science, Master of Science - MSc, Data Science at ETH ZürichExchange Semester, Exchange Semester at Delft University of Technology
German, English, French, Danish
Github Skills (24)
visualization10
electricity10
consumption10
sustainability10
climate-change10
electricity-prices10
data-visualization10
representation-learning6
gnn5
temporal4
interaction4
embedding3
machine-learning3
graph3
pytorch2
Programming languages (1)
Github contributions (5)
florianscheidl/k-gnn
Oct 2022 - Oct 2022
Source code for our AAAI paper "Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks".
Contributions:113 pushes, 1 branch in 10 days
pytorchdeep-learningneural-graphneural-networksmachine-learning
florianscheidl/pytorch_geometric
Aug 2022 - Jan 2023
Graph Neural Network Library for PyTorch
Contributions:152 pushes, 2 branches in 4 months
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
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.