Krzysztof Łęcki
Senior Software Development Engineer at NVIDIA
Warsaw, Masovian Voivodeship, Poland
Summary
🤩
RockstarKrzysztof Łęcki is a Senior Software Development Engineer based in Warsaw with seven years of experience building high-performance backend systems for deep learning workflows. At NVIDIA since 2020, he contributes to the prominent NVIDIA/DALI project, focusing on low-level image processing, CUDA memory handling, and efficient data augmentation operators that accelerate training pipelines. He combines practical bug-fixing and API hardening with implementing optimized CPU and GPU operators, demonstrating both systems-level rigor and attention to developer-facing reliability. Known for tackling subtle issues like BMP parsing and CUDA copy semantics, he brings a pragmatic, detail-oriented approach to performance-critical code.
7 years of coding experience
Polish, English, szwedzki
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (14)
cuda10
c-language10
gpgpu10
cprogramming-language10
gpu10
image-processing10
accelerated-computing10
optimization9
test-automation9
optimisation9
numerical-optimization9
algorithm9
code-optimization9
algorithms9
Programming languages (6)
Github contributions (5)
NVIDIA/DALI
Sep 2018 - Jan 2023
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:
Backend Developer
Backend DeveloperContributions:20 releases, 2434 reviews, 388 commits in 4 years 5 months
Contributions summary:Krzysztof primarily worked on the DALI library, focusing on low-level image processing and deep learning data pipelines. Their contributions involved fixing bugs related to BMP image size reading and CUDA-related memory and copy operations. The user also implemented and refined various operators, including a CPU-based Gaussian blur and a set of mathematical expressions used in data augmentation and processing, demonstrating a strong understanding of the library's core functionalities. The code changes include general improvements of error handling in several API functions.
pythontrainingpaddlegpu-accelerationbuilding-blocks
klecki/test_repo
Apr 2019 - Nov 2021
Test repository
Contributions:4 PRs, 2 pushes, 4 branches in 2 years 7 months
testingtest-repository
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.