Mathias Gruber
Chief Data Scientist
Copenhagen, Capital Region of Denmark
Summary
🤩
Rockstar🎓
Top SchoolMathias Gruber is a Chief Data Scientist based in Copenhagen with 11+ years applying advanced machine learning and deep learning across pharma, biotech and industrial settings. He builds production-ready MLOps platforms and user-facing AI products (Django/React/Serverless) while leading projects from Bayesian optimization for chemistry to GNNs for drug discovery and RAG/LLM systems for compliant document retrieval. His background in nanoscience and a PhD in biophysics informs a pragmatic, scientific approach to uncertainty estimation and interpretability in models used for regulated industries. He has strong hands-on ML engineering experience—evident from open-source work such as implementing mask generation and prediction utilities for a partial-convolution image inpainting Keras repo—and a track record of shipping scalable solutions that bridge research and product. Outside corporate roles he founded and ran a profitable online game with half a million registered users, reflecting a long-standing ability to design and operate high-traffic systems.
11 years of coding experience
3 years of employment as a software developer
Master of Science Nanotechnology, Master of Science Nanotechnology at Københavns Universitet - University of CopenhagenNanodegree Deep Reinforcement Learning, Nanodegree Deep Reinforcement Learning at Udacity
English, Danish, German
Stackoverflow
1reputation
0reached
1answer
0questions
Github Skills (10)
keras10
computer-vision10
opencv10
python10
tensorflow9
deep-learning9
machine-learning8
neural-network6
classification6
multiclass-classification6
Programming languages (6)
Github contributions (5)
MathiasGruber/PConv-Keras
Jun 2018 - Aug 2021
Unofficial implementation of "Image Inpainting for Irregular Holes Using Partial Convolutions". Try at: www.fixmyphoto.ai
Role in this project:
ML Engineer
ML EngineerContributions:52 commits, 29 pushes, 1 branch in 3 years 2 months
Contributions summary:Mathias implemented the initial mask generation functionality for the image inpainting project. They added a notebook to explain the process of generating irregular masks using OpenCV, referencing a relevant research paper. Further contributions included modifications to the prediction process, introducing the ability to specify batch sizes, and adding a convenience function to handle predictions on arbitrary image sizes. The user also added functionality to load VGG16 weights.
pytorchholespythonpartialnvidia
MathiasGruber/TheNinjaRPG
Jul 2023 - Apr 2025
Source code for TheNinja-RPG
Contributions:69 reviews, 139 PRs, 2134 pushes in 1 year 9 months
aigamet3
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.