Johanna Sommer
Machine Learning Research Engineer at Pruna AI
Munich, Bavaria, Germany
Summary
🤩
Rockstar🎓
Top SchoolJohanna Sommer is a Machine Learning Research Engineer and PhD candidate at TU Munich with nine years of experience bridging industrial research and academic ML. She progressed from a dual study program and multiple IBM research internships—working on AutoML, meta-learning for gradient boosting, and large-scale sparse matrix techniques—to research roles at TUM and now Pruna AI. Her work blends hands-on systems thinking with theoretical rigor, often focusing on efficient, scalable ML algorithms and practical deployment considerations. Notably, she contributed to Apache SystemML during an IBM Almaden internship, signaling early commitment to impactful open-source and large-scale ML. Based in Munich, she aims to translate doctoral research into production-ready solutions that improve model selection and training efficiency.
9 years of coding experience
3 years of employment as a software developer
Bachelor of Applied Science (BASc) Applied Computer Science, Bachelor of Applied Science (BASc) Applied Computer Science at Baden-Wuerttemberg Cooperative State University (DHBW)Doctor of Philosophy - PhD Machine Learning, Doctor of Philosophy - PhD Machine Learning at Technical University of Munich
German, English, Italian, French, Arabic
Github Skills (42)
transformers10
python10
diffusion-models10
machine-learning10
combinatorial-optimization10
dml10
diffusers10
java10
operations-research10
llm10
deep-learning10
ai10
flux10
computer-vision10
optimization10
Programming languages (6)
Github contributions (5)
johannaSommer/generalization-neural-co-solvers
Jan 2022 - Nov 2022
Official Repository for the ICLR 2022 paper "Generalization of Neural Combinatorial Solvers through the Lens of Adversarial Robustness"
Contributions:10 commits, 3 PRs, 8 pushes in 9 months
pytorchiclrrobustnessgeneralizationdeep-learning
PrunaAI/pruna
Mar 2025 - Feb 2026
Pruna is a model optimization framework built for developers, enabling you to deliver faster, more efficient models with minimal overhead.
Contributions:3 releases, 141 reviews, 126 PRs in 11 months
aicomputer-visiondeep-learningllmmachine-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.