Christopher Brix
PhD Candidate
Aachen, North Rhine-Westphalia, Germany
Summary
👤
Senior🎓
Top SchoolChristopher Brix is a PhD candidate at RWTH Aachen with 11 years of software and ML experience focused on making neural networks verifiable and more secure. He combines deep research skills with hands-on engineering from internships at AWS and Amazon—where he replicated and extended AlphaDev with formal verification and benchmarked LLM robustness across 10,000+ trained models—to Google work that boosted NER metrics by up to 25 percentage points. His open-source contributions span practical ML tooling and RNN training frameworks (notably adding 2D-LSTM support in returnn) and educational notebooks for ML summer schools, reflecting both production-ready code and clear pedagogy. Based in Aachen, he brings a rare mix of formal methods, empirical robustness evaluation, and NLP expertise that targets real-world AI reliability challenges.
11 years of coding experience
4 years of employment as a software developer
Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at RWTH Aachen University
Github Skills (13)
machine-learning10
jupyter-notebook10
recurrent-neural-networks10
deep-learning10
tensorflow10
python10
natural-language-processing9
testing9
reinforcement-learning9
gpu8
scikit7
scikit-learn7
hdf7
Programming languages (9)
Github contributions (5)
rwth-i6/returnn
Dec 2018 - Nov 2019
The RWTH extensible training framework for universal recurrent neural networks
Role in this project:
ML Engineer
ML EngineerContributions:9 commits, 16 PRs, 53 comments in 11 months
Contributions summary:Christopher's commits primarily focused on modifying and testing the `TFNetworkRecLayer` within the `returnn` repository, an extensible training framework for recurrent neural networks. They implemented and tested new features, specifically adding support for 2D-LSTM. The user also added functionality to handle constraints and ensure subnetwork layers respect the `trainable` flag, contributing to the framework's flexibility and control over model training. Additionally, the user fixed an HDFDumpLayer issue, indicating a focus on data handling within the network.
recurrent-neural-networksgputensorflowtheanodeep-learning
LxMLS/lxmls-toolkit
Jul 2019 - Jul 2019
Machine Learning applied to Natural Language Processing Toolkit used in the Lisbon Machine Learning Summer School
Role in this project:ML Engineer
ML EngineerContributions:10 commits, 10 PRs, 7 comments in 2 days
Contributions summary:Christopher primarily focused on updating and maintaining the Jupyter Notebook tutorials within the repository, adding "magic reload" commands to notebooks, and removing unnecessary iterations. Their work modified several notebooks related to machine learning topics such as linear classifiers, reinforcement learning, and sequence models. They also made minor style fixes and replaced an unknown character.
machine-learningnatural-language-processing
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.