Michal Szutenberg
Performance Engineer at (undisclosed)
Gdansk, Pomeranian Voivodeship, Poland
Summary
🤩
Rockstar🎓
Top SchoolMichal Szutenberg is a performance engineer with seven years of experience specializing in optimizing large language model inference and ML benchmarking. He has driven MLPerf submissions and designed efficient TPC kernels and inference stacks while working across Intel and Habana Labs, focusing on vLLM, TGI, and Gaudi accelerator support. Michal blends low-level firmware and kernel work with infrastructure design—building internal benchmarking tooling, graph optimizations for TensorFlow, and developer automation like qnpu and Jenkins initiatives. He mentors engineers and contributes to industry benchmarking efforts (MLCommons) where he helped shape LLM inference benchmarks for models like GPT-J and LLaMA-70B. Based in Gdansk, he pairs academic training in embedded systems with practical experience from embedded firmware to cutting-edge deep learning performance engineering. A less obvious strength is his recurring role as a bridge between product customers and low-level implementation, translating real workloads into concrete optimization wins.
7 years of coding experience
8 years of employment as a software developer
Master of Science - MS Embedded Systems, Master of Science - MS Embedded Systems at Eindhoven University of TechnologyMaster of Science - MS ICT Innovation, Master of Science - MS ICT Innovation at KTH Royal Institute of TechnologyBachelor of Engineering (BEng) Electronics and Telecommunications, Bachelor of Engineering (BEng) Electronics and Telecommunications at Gdańsk University of Technology
Polish, English, Russian
Github Skills (32)
transformers10
pytorch10
benchmark10
language-model10
python10
data-science10
tensorflow-addons10
machine-learning10
benchmarking10
seq2seq10
flax10
ml10
deep-learning10
tensorflow10
deep-neural-networks10
Programming languages (4)
Github contributions (5)
szutenberg/tensorflow
Mar 2021 - May 2022
An Open Source Machine Learning Framework for Everyone
Contributions:15 pushes, 10 branches in 1 year 3 months
pythondata-sciencedeep-learningmachine-learningframework-learning
szutenberg/TUeLaneTracker2
Jan 2019 - Jul 2019
Contributions:119 pushes, 2 branches in 6 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.