Stefan Grafberger
Machine Learning Engineer at BIFOLD & TU Berlin
Berlin, Germany
Summary
🤩
RockstarStefan Grafberger is a Machine Learning Engineer and Ph.D. student at BIFOLD & TU Berlin with a decade of experience working at the intersection of data management and machine learning. Based in Berlin, he combines research rigor with hands-on backend development and test automation, contributing practical improvements to large-scale data quality tooling like AWS Labs' Deequ. His contributions—such as richer constraint hints, streamlined suggestion APIs, and more robust tests—demonstrate a focus on making ML pipelines more reliable and interpretable. Stefan brings an engineer’s mindset to research problems, turning theoretical insights into usable code and developer-friendly APIs.
10 years of coding experience
German, English
Github Skills (7)
unit-testing10
spark10
data-quality10
scala10
test-automation9
apache-spark9
data-analysis8
Programming languages (7)
Github contributions (5)
awslabs/deequ
Sep 2018 - Sep 2018
Deequ is a library built on top of Apache Spark for defining "unit tests for data", which measure data quality in large datasets.
Role in this project:
Back-end Developer & Test Automation Engineer
Back-end Developer & Test Automation EngineerContributions:14 commits, 8 PRs, 32 pushes in 16 days
Contributions summary:Stefan primarily contributed to the Deequ library by adding functionality and improving the existing code. They introduced a hint parameter to constraints, enhancing error messages, and streamlined the Constraint Suggestion API, adding features and improving the structure of its results. Additionally, they fixed a test that could fail on rare occasions and cleaned up code, particularly focusing on constraint suggestions and column profiles. Their work included adding convenience functions and new constraint suggestion rules.
data-qualityapacheunitsparkunit-tests
stefan-grafberger/mlinspect
Jul 2020 - Nov 2022
Inspect ML Pipelines in Python in the form of a DAG
Contributions:6 reviews, 168 commits, 45 PRs in 2 years 3 months
pythonml-pipelinesdata-sciencedaginspect
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.