Maciej Kula
Senior Staff Software Engineer at Google DeepMind
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolMaciej Kula is a Senior Staff Software Engineer with 13 years of experience building large-scale recommender systems and applied ML, currently focused on LLM post-training, tool use, and synthetic data at Google DeepMind. He previously led recommender research and infrastructure across YouTube, Google Play, Google Search and Ads, and drove personalization at Netflix and Lyst. A hands-on engineer and researcher, he has contributed to prominent open-source recommender projects (including Spotlight and LightFM) and to TensorFlow Recommenders, often implementing core C/Cython/PyTorch optimizations. His background in economics (MPhil, Oxford) and early work in economic consulting underpin a measured, data-driven approach to modeling and evaluation. Colleagues rely on him to bridge research and production—turning novel ranking and representation ideas into scalable systems. He’s quietly known for squeezing big performance wins from low-level code while keeping product and user metrics front and center.
13 years of coding experience
8 years of employment as a software developer
M.Phil., Economics, M.Phil., Economics at University of OxfordRecsys 2013
Github Skills (22)
algorithm10
algorithms10
pytorch10
python10
machine-learning10
c1110
data-structure10
numpy10
c1710
matrix-factorization10
deep-learning10
tensorflow10
cython10
recommender-system10
data-structures10
Programming languages (16)
Github contributions (5)
maciejkula/glove-python
Oct 2014 - May 2016
Toy Python implementation of http://www-nlp.stanford.edu/projects/glove/
Role in this project:
Back-end Developer & Algorithm Implementer
Back-end Developer & Algorithm ImplementerContributions:88 commits, 26 PRs, 34 pushes in 1 year 7 months
Contributions summary:Maciej primarily focused on optimizing and extending the core functionality of the GloVe model. They implemented a Cython-based co-occurrence matrix construction, significantly speeding up the process. Furthermore, they added code to compute paragraph vector representations, enhancing the model's capabilities. The user also contributed to the project's installability and added a model evaluation routine, demonstrating an understanding of the project's overall functionality.
python
tensorflow/recommenders
Jun 2020 - Nov 2022
TensorFlow Recommenders is a library for building recommender system models using TensorFlow.
Role in this project:Back-end Developer
Back-end DeveloperContributions:20 releases, 12 reviews, 134 commits in 2 years 4 months
Contributions summary:Maciej contributed to the TensorFlow Recommenders library, focusing on adding features and improving test coverage. Their work includes adding version information to the initialization file, adding test cases, and refactoring base model functions. They also added a testing script for example notebooks and made internal changes to improve the code.
recommender-systemtensorflowtensorflow-recommendersrecommender
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.