Ladislav Rampasek
Senior Research Scientist at Isomorphic Labs
London, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolLadislav Rampasek is a Senior Research Scientist at Isomorphic Labs with 14 years of experience applying geometric deep learning and probabilistic modeling to biomedical problems. He earned a PhD from the University of Toronto and completed postdoctoral work in prominent geometric deep learning groups at Mila and Université de Montréal, focusing on latent-variable models and variational inference for cancer treatment prediction and drug perturbation effects. His background spans industry internships at Google and Atomwise and contributions to open-source graph transformer tooling—recently updating GraphGPS training and checkpointing for PyG v2 compatibility to streamline fine-tuning and deployment. Based in London, he blends rigorous academic research with practical engineering, optimizing model training pipelines for real-world drug discovery applications. Known for translating complex probabilistic methods into scalable code, he brings both theoretical depth and hands-on systems experience to translational AI projects.
14 years of coding experience
7 years of employment as a software developer
Master's Degree Computer Science, Master's Degree Computer Science at Univerzita Komenského v BratislaveDoctor of Philosophy (Ph.D.) Computer Science, Doctor of Philosophy (Ph.D.) Computer Science at University of Toronto
Github Skills (7)
machine-learning10
pytorch10
graph-neural-network10
python10
model-checking9
checkpointing9
checkpoint9
Programming languages (5)
Github contributions (5)
rampasek/GraphGPS
May 2022 - Jan 2023
Recipe for a General, Powerful, Scalable Graph Transformer
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2 releases, 16 commits, 18 PRs in 7 months
Contributions summary:Ladislav refactored and updated component registration for compatibility with PyGv2.0.4, focusing on optimizer and scheduler configurations. They also modified the main training script to integrate pretrained models and fine-tuning capabilities, which involved changes to the training and inference pipelines. Furthermore, the user modified model checkpointing and loading mechanisms. These changes indicate a focus on model training and deployment, optimizing the existing code for the updated PyG version.
graphgraph-representation-learninglong-range-dependencegraph-neural-networkgraph-transformer
rampasek/nn
Aug 2016 - Nov 2021
Contributions:2 pushes in 5 years 4 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.