Kinga Gajdamowicz
Deep Learning Software Engineer at Intel Corporation
Gdańsk, Pomeranian Voivodeship, Poland
Summary
🤩
Rockstar🎓
Top SchoolKinga Gajdamowicz is a Deep Learning Software Engineer based in Gdańsk with six years of experience building high-performance ML and distributed training infrastructure at Intel. She focuses on backend and performance engineering for graph neural network tooling, contributing to PyTorch Geometric by improving global pooling, adding distributed and temporal/edge-level sampling, and enhancing benchmark capabilities and CSV export for reproducible performance testing. Her background in electronics and telecommunications plus a master's in computer science gives her a pragmatic systems view that spans from low-level networking to large-scale deep learning workloads. Known for optimizing distributed sampling strategies, she combines production-grade engineering with research-oriented problem solving to accelerate graph-based model training.
6 years of coding experience
2 years of employment as a software developer
Magister (Mgr), Computer Science, Magister (Mgr), Computer Science at Politechnika Gdańska
English, Polish
Github Skills (10)
pytorch10
benchmark10
benchmarking10
distributed-training10
graph-neural-network10
performance-optimization10
python10
graph-convolutional-networks9
deep-learning9
geometric-deep-learning9
Programming languages (2)
Github contributions (5)
pyg-team/pytorch_geometric
Nov 2022 - Jan 2023
Graph Neural Network Library for PyTorch
Role in this project:
Backend & Performance Engineer
Backend & Performance EngineerContributions:48 reviews, 2 commits, 18 PRs in 2 months
Contributions summary:Kinga primarily focused on enhancing the benchmarking capabilities of the PyTorch Geometric library. Their contributions included enabling the setting of custom step counts within benchmarks, and adding functionality to save benchmark results to CSV files. They also worked on improving the performance of global pooling operations and implemented distributed sampling strategies, specifically addressing temporal and edge-level sampling for both homogeneous and heterogeneous graphs, showcasing a focus on distributed training optimization.
pytorchgraph-convolutional-networksgeometric-deep-learningdeep-learningneural-graph
kgajdamo/pyg-lib
Jul 2022 - Mar 2024
Low-Level Graph Neural Network Operators for PyG
Contributions:173 pushes, 41 branches, 5 comments in 1 year 8 months
graph-convolutional-networksgeometric-deep-learningpygneural-graphgraph-neural-network
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.