Paul Voigtlaender
Senior Research Scientist at Google DeepMind
Aachen, North Rhine-Westphalia, Germany
Summary
👤
Senior🎓
Top SchoolPaul Voigtlaender is a senior research scientist at Google DeepMind with 14 years of experience building multimodal AI systems and advancing spatial and localized understanding in state-of-the-art models like Gemini and Nano Banana. He has a strong research-to-production track record at Google Research and DeepMind, from video-language grounding and Video Localized Narratives to leading conversational image segmentation across flagship multimodal models. Paul holds a PhD in Computer Science from RWTH Aachen and contributes to influential open-source tooling for sequence models, including backend work on the RWTH returnn framework. Colleagues rely on him to bridge deep research insights with engineering rigor, debugging and refactoring complex codepaths to make novel model capabilities usable in practice.
14 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at RWTH Aachen University
Github Skills (14)
deep-q-learning10
recurrent-neural-networks10
deep-learning10
theano10
python9
machine-learning9
c-language8
debug8
debugging8
cprogramming-language8
tensorflow8
apidoc7
api7
develop7
Programming languages (1)
Github contributions (5)
rwth-i6/returnn
Jan 2014 - Oct 2018
The RWTH extensible training framework for universal recurrent neural networks
Role in this project:
Back-end Developer
Back-end DeveloperContributions:201 commits, 28 pushes, 78 comments in 4 years 9 months
Contributions summary:Paul primarily contributed to the RWTH extensible training framework for universal recurrent neural networks. They made changes focusing on core functionalities of the framework, including modifications to the underlying support code and addition of features related to sequence training. Their work demonstrates an understanding of the codebase and its interaction with lower-level libraries. The user's commits point towards debugging and refactoring efforts, particularly concerning indentation and compatibility issues within different modules.
rwthdeep-learninggpurecurrent-neural-networkstheano
pvoigtlaender/got10k-toolkit
Sep 2019 - Apr 2020
Official Python toolkit for generic object tracking benchmark GOT-10k and beyond
Contributions:22 commits, 22 pushes in 6 months
pythongenerictrackingbenchmarkpython-toolkit
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.