Krystian Sztenderski
Software Engineer at Google
Warsaw, Masovian Voivodeship, Poland
Summary
🤩
Rockstar🎓
Top SchoolKrystian Sztenderski is a software engineer with six years of experience specializing in ML-enabled backend systems and data pipelines, currently working at Google after roles at Nomagic and an internship at NVIDIA. He contributed to NVIDIA DALI—an influential GPU-accelerated data loading library—adding robust file filtering, glob matching, and a prototype debug mode that show practical depth in I/O, performance-sensitive code, and developer tooling. Krystian blends machine learning engineering and low-level systems work, having moved from early internships at Samsung and NVIDIA into production ML roles. Based in Warsaw and pursuing an MS at the University of Warsaw, he pairs academic grounding with hands-on open-source contributions that improve real-world ML training workflows.
6 years of coding experience
3 years of employment as a software developer
Master of Science - MS, Master of Science - MS at University of Warsaw
Github Skills (17)
filesystem10
c-language10
data-pipelines10
python10
machine-learning10
file-operations10
deep-learning10
fileio10
data-processing10
cprogramming-language10
data-pipeline10
debug9
debugging9
gpu9
data-augmentation9
Programming languages (5)
Github contributions (5)
NVIDIA/DALI
Nov 2021 - Jun 2022
A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:187 reviews, 20 commits, 38 PRs in 7 months
Contributions summary:Krystian primarily contributed to the development of the DALI (Data Loading Library) framework. Their work focused on enhancing the data processing capabilities of the library, specifically by implementing file filtering functionalities in the FileReader operator. They added support for glob pattern matching and case-sensitivity control for file selection, demonstrating an understanding of data loading pipelines and file system operations. The user also developed a prototype for a debug mode, including features like data access during pipeline execution and arithmetic operations for debugging.
pythontrainingpaddlegpu-accelerationbuilding-blocks
ksztenderski/DALI
Oct 2021 - Jun 2022
A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.
Contributions:144 pushes, 18 branches in 8 months
traininggpu-accelerationbuilding-blocksexecution-enginegpu
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.