Alexis Duburcq
Member Of Technical Staff at Genesis AI
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolAlexis Duburcq is a robotics and software engineer with a decade of experience building control systems, simulators, and reinforcement learning pipelines for humanoid robots and exoskeletons. He led development of Jiminy, a high-performance Python/C++ simulator compatible with OpenAI Gym, and contributed core data-structure refactors to the popular tianshou RL library, improving NumPy/PyTorch interoperability and replay buffers. His work spans research and product: a PhD-level background in computer science and top robotics degrees from EPFL and Centrale Paris inform fast, memory‑efficient C++ implementations and real-time control validated on human trials. As a co-founder and CTO turned Member of Technical Staff, he combines entrepreneurial drive with hands-on systems engineering across simulation, optimization, and ML. Notably, his master’s thesis cut gait optimization time from an hour to two minutes via analytical reduction and code optimization—evidence of his focus on making advanced algorithms practical and reliable. Based in Paris, he excels at interdisciplinary collaboration that turns complex dynamics into deployable robotic behaviors.
10 years of coding experience
8 years of employment as a software developer
Engineer’s Degree (Centrale Paris), Computer Science and Advanced Systems, 3.93/4.33, Engineer’s Degree (Centrale Paris), Computer Science and Advanced Systems, 3.93/4.33 at Ecole Centrale ParisMaster’s Degree, Robotics and autonomous systems, 5.39/6, Master’s Degree, Robotics and autonomous systems, 5.39/6 at Ecole polytechnique fédérale de LausanneDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Université Paris Dauphine - PSL
French, English, Spanish
Stackoverflow
1,054reputation
320kreached
55answers
1question
pytorch
top-5%
Github Skills (11)
data-structures10
pytorch10
python10
batch-processing10
data-structure10
reinforcement-learning10
numpy9
hdf6
python-dataclasses6
convex-hull6
h5py6
Programming languages (12)
Github contributions (5)
thu-ml/tianshou
May 2020 - Jul 2020
An elegant PyTorch deep reinforcement learning library.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:162 reviews, 21 commits, 21 PRs in 1 month
Contributions summary:Alexis primarily focused on refactoring and improving the `Batch` class, the core data structure of the project. This included addressing issues related to data conversion between NumPy and PyTorch, and ensuring that the data structure could handle the diverse data used in reinforcement learning. The user also worked on enhancing the replay buffer and other data structures, adding features and fixing bugs to improve the robustness and functionality of the library. These changes demonstrate a focus on maintaining and improving core components of the reinforcement learning library.
deep-reinforcement-learningbenchmarknpgtd3reinforcement
duburcqa/Genesis
Feb 2025 - Apr 2025
A generative world for general-purpose robotics & embodied AI learning.
Contributions:1 review, 5 PRs, 293 pushes in 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.