Maria Samsikova

Research Engineer at Google DeepMind

United Kingdom
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
🎓
Top School
Maria Samsikova is a research engineer with seven years of experience blending applied mathematics and software engineering, currently contributing to DeepMind’s research efforts in the UK. She has a track record of internships and engineering roles at Google, Yandex, and JetBrains, progressing from STEP and SWE internships to a full research engineer position. Maria contributes to high-impact open-source ML tooling—most notably improving loss functions in DeepMind’s Optax library for JAX, adding KL-divergence support and numerical underflow protections. Her work sits at the intersection of optimization, reliable numerical implementation, and reproducible research, reflecting strong applied-math instincts from her ITMO University background. Colleagues describe her as detail-oriented in core algorithmic code and pragmatic about production-quality documentation and tests.
code7 years of coding experience
job1 year of employment as a software developer
bookBachelor of Science - BS, Applied Mathematics and Computer Science, Bachelor of Science - BS, Applied Mathematics and Computer Science at ITMO University
languagesGerman, Russian, English
github-logo-circle

Github Skills (6)

machine-learning10
jax10
python10
optimization10
unit-testing9
numpy8

Programming languages (5)

JavaTeXJavaScriptJupyter NotebookPython

Github contributions (5)

github-logo-circle
google-deepmind/optax

Feb 2022 - Jun 2022

Optax is a gradient processing and optimization library for JAX.
Role in this project:
userML Engineer
Contributions:10 reviews, 12 commits, 3 PRs in 3 months
Contributions summary:Maria primarily focused on implementing and refining loss functions within the Optax library. They added the Kullback-Leibler divergence loss, corrected documentation, and improved the loss function by adding underflow protection and refactoring it into two distinct functions. The contributions involved modifying and testing core machine learning components related to optimization and loss calculations.
deep-learningoptimizationmachine-learningoptaxjax
holounic/university

Sep 2019 - Oct 2022

Contributions:171 commits, 124 pushes, 3 comments in 3 years 1 month
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.
Request Free Trial
Maria Samsikova - Research Engineer at Google DeepMind