Sertan Girgin
Software Engineer at Google DeepMind
Greater Paris Metropolitan Region France
Summary
🤩
Rockstar🎓
Top SchoolSertan Girgin is a software engineer with six years of industry experience and a deep research pedigree, currently contributing to Google DeepMind from the Greater Paris area. He has a long tenure across Google and YouTube and earlier research roles at INRIA and universities, blending production-grade engineering with academic rigor from a Ph.D. in Computer Engineering. His open-source work on high-profile DeepMind projects like Acme and OpenSpiel shows practical expertise in reinforcement learning, mean field game algorithms, and robust refactoring of core components and testing. Sertan’s contributions often focus on improving algorithmic correctness, data handling, and hybrid offline/online agent support—skills that bridge ML research and backend systems. Colleagues describe him as someone who moves fluidly between research code and production constraints, able to translate novel algorithms into maintainable libraries. He brings a quiet, methodical approach informed by years of academic research and large-scale engineering practice.
6 years of coding experience
19 years of employment as a software developer
Ph.D. Computer Enginnering, Ph.D. Computer Enginnering at Orta Doğu Teknik Üniversitesi / Middle East Technical University
English, Turkish, French
Github Skills (15)
algorithm10
coregames10
algorithms10
machine-learning10
multi-agent10
c-language10
jax10
cprogramming-language10
python10
reinforcement-learning10
videogames10
tensorflow9
agent9
testing8
openai-gym7
Programming languages (4)
Github contributions (5)
google-deepmind/acme
Nov 2021 - Aug 2022
A library of reinforcement learning components and agents
Role in this project:
ML Engineer
ML EngineerContributions:8 commits in 9 months
Contributions summary:Sertan primarily contributed to the reinforcement learning components of the Acme library. They refactored game logic by changing checks for sequential games, and fixed type casting issues within the GymWrapper related to rewards. Further contributions included unit tests for the episode_to_timestep_batch method and refactoring for hybrid offline/online agent support, indicating a focus on improving the library's core functionality and data handling.
reinforcement-learningreinforcementagentsdeep-reinforcement-learning
google-deepmind/open_spiel
Sep 2021 - Nov 2022
OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.
Role in this project:Back-end Developer
Back-end DeveloperContributions:14 reviews, 36 commits, 20 comments in 1 year 1 month
Contributions summary:Sertan primarily focused on updating and refactoring the Mean Field Game (MFG) algorithms within the `open_spiel` repository. They modified existing algorithms to utilize value functions and added new methods related to distribution manipulation. These changes involved refactoring existing code related to policy calculations, best response implementation, and more. Furthermore, the user added examples and utilities, particularly related to the DQN and mirror descent methods, demonstrating expertise in MFG algorithms.
cppmultiagentgamespythondatamining
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.