Darius Morawiec
Senior Data Scientist
Germany
Summary
🤩
Rockstar🎓
Top SchoolDarius Morawiec is a Senior Data Scientist with 11 years of experience building and shipping scalable ML services at OTTO, spanning data engineering, model training, deployment and observability on GCP with Docker and Terraform. He combines hands-on expertise in computer vision and multimodal models (classification, regression, object detection) with production-grade engineering—optimizing inference, managing ANNOY/approximate kNN indices and designing realtime pipelines. Comfortable across the stack, he prototypes APIs and UIs (FastAPI, Vue.js), builds ETL and labeling tools, and turns model evaluations into stakeholder-facing reports and Power BI dashboards. An early maker with a security-champion mindset, he supervises trainees and embeds governance into ML systems. He is an active open-source contributor, improving tools like sklearn-porter to transpile scikit-learn models into multiple languages and has front-end performance contributions to p5.js legacy site. Based in Germany, he continually experiments with new frameworks and languages to close the gap between research prototypes and robust production services.
11 years of coding experience
5 years of employment as a software developer
Master of Science (M.Sc.) Computer Science and Media, Master of Science (M.Sc.) Computer Science and Media at Stuttgart Media University
English, German
Github Skills (17)
javascript10
python10
data-science10
scikit10
css10
machine-learning10
gruntjs10
web-performance10
scikit-learn10
g10
optimization10
data-structure9
algorithm9
algorithms9
postcss9
Programming languages (17)
Github contributions (5)
nok/sklearn-porter
Aug 2016 - May 2022
Transpile trained scikit-learn estimators to C, Java, JavaScript and others.
Role in this project:
Data Scientist
Data ScientistContributions:15 releases, 1160 commits, 15 PRs in 5 years 10 months
Contributions summary:Darius made multiple code contributions to the `sklearn-porter` repository, focused on enhancing its capabilities. The commits indicate work on extending the tool's ability to predict a set of features, iterating on arrays, and improving code quality. Additionally, they added full comments to the `Algorithm.py` file and renamed the template class.
transpilepythondata-sciencesklearnjavascript
processing/p5.js-website-legacy
Aug 2016 - Nov 2016
Archived p5.js website 2015-2024
Role in this project:Front-end Developer
Front-end DeveloperContributions:32 commits, 10 PRs, 14 comments in 3 months
Contributions summary:Darius primarily contributed to optimizing the p5.js website by adding image and CSS optimizations. They implemented image compression using Grunt, concatenated and minified CSS files, and applied CSS optimizations using PostCSS. The user also introduced JavaScript optimizations, including code minification and the addition of async non-blocking JavaScript handling, improving the website's performance and user experience.
gruntnode-jsjavascriptp5-jsnodejs
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.