Piotr Marcinkiewicz
Machine Learning Engineer at Nvidia
Warsaw, Masovian Voivodeship, Poland
Summary
🤩
RockstarPiotr Marcinkiewicz is a Machine Learning Engineer based in Warsaw with five years of hands-on experience building and deploying production ML systems. At NVIDIA he contributes to NeMo Framework Launcher, focusing on efficient inference for chatbots and text generation, including integration with FasterTransformer and slurm executors for scalable on-prem and cloud deployments. He combines practical engineering—vocabulary validation, configuration querying, and README-driven documentation improvements—with performance-minded implementation and testing. Piotr’s work bridges model development pipelines and inference optimization, helping teams move large language and multimodal models from prototype to production. Colleagues describe him as a pragmatic problem-solver who pays attention to both developer UX and runtime efficiency.
5 years of coding experience
Github Skills (9)
inference10
python10
ml-deployment9
machine-learning9
triton-server9
nlp9
continuous-deployment9
triton9
gpt8
Programming languages (5)
Github contributions (5)
NVIDIA/NeMo-Framework-Launcher
Nov 2021 - Oct 2022
Provides end-to-end model development pipelines for LLMs and Multimodal models that can be launched on-prem or cloud-native.
Role in this project:
ML Engineer
ML EngineerContributions:77 commits, 1 comment in 11 months
Contributions summary:Piotr primarily contributed to the implementation and testing of an inference chatbot. Their work involved creating a text generation system and implementing chatbot functionality, likely using the FasterTransformer library for efficient inference. They integrated vocabulary validation and configuration querying within the inference workflow. The user also addressed slurm executor parameters and updated the README file.
piotrm-nvidia/pytriton
Aug 2024 - Dec 2024
PyTriton is a Flask/FastAPI-like interface that simplifies Triton's deployment in Python environments.
Contributions:1 PR, 12 pushes, 1 branch in 4 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.