Bartłomiej Cieślar
PHD Student at Imperial College London
Poland
Summary
🤩
Rockstar🎓
Top SchoolBartłomiej Cieślar is a PhD student and software engineer with eight years of experience at the intersection of deep learning infrastructure, compilers, and systems engineering. He has contributed production-grade features to high-profile projects and companies—implementing a performant webdataset reader for NVIDIA DALI, adding CUDA support to Clang-tidy at Meta, and prototyping GHC features during a Haskell DevX internship at IOHK. His practical background spans large-model bringup on Tenstorrent hardware, GPU-accelerated data pipelines, and applied neural networks, complemented by hands-on 3D modeling and printing. Educated at Imperial College London and an exchange stint at MIT, he pairs rigorous academic training with competitive algorithmic problem-solving. Colleagues describe him as a pragmatic engineer who closes the loop from research ideas to optimized, tested implementations. He’s based in Poland and brings an unusual blend of low-level systems optimization and machine learning deployment experience.
8 years of coding experience
Exchange Computer Science, Exchange Computer Science at Massachusetts Institute of TechnologyMaster of Computer Science Computer Science, Master of Computer Science Computer Science at Imperial College LondonStudent Mathematics and Computer Science, Student Mathematics and Computer Science at XIV Liceum Ogólnokształcące im. Stanisława Staszica w WarszawieMiddle School Education, Middle School Education at 58th Middle School of Władysław IV in Warsaw
German, Polish, Spanish, English
Github Skills (14)
data-loading10
python10
haskell10
data-processing10
compiler-development10
ghc10
abstract-syntax-tree9
cprogramming-language9
performance-optimization9
c-language9
computer-engineering8
gpu7
automations5
automation5
Programming languages (9)
Github contributions (5)
NVIDIA/DALI
Jul 2021 - Oct 2021
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 & DevOps Engineer
Back-end Developer & DevOps EngineerContributions:213 reviews, 12 commits, 24 PRs in 2 months
Contributions summary:Bartłomiej's primary contributions revolve around the implementation of a webdataset reader for the DALI library, enabling data loading from webdataset archives. They added bindings and implemented the API for libtar, including associated tests and code cleanup. Furthermore, they optimized the webdataset reader for performance improvements and added automatic index file inference.
pythontrainingpaddlegpu-accelerationbuilding-blocks
OliverKillane/Arm-11-Group-Project
May 2021 - Aug 2021
Contributions:145 commits in 3 months
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.