Cournapeau David
Engineering Manager, Applied Science at Amazon
Chiyoda, Japan
Summary
👤
Senior🎓
Top SchoolCournapeau David is an engineering leader with 19 years of experience combining scientific computing, machine learning and large-scale production systems, currently managing Applied Science at Amazon. He founded scikit-learn and is a former core contributor to NumPy and SciPy, bringing rare depth in C/Python internals and numerical correctness to ML infrastructure. At Mercari he built and scaled cross-functional organizations delivering real-time ML that measurably increased search conversion while managing multi-million-dollar cloud budgets. He has repeatedly built teams from scratch—from research-oriented ML groups to production engineering orgs—and shifted cultures toward reproducible MLOps, systematic evaluation, and accountable roadmaps. Equally comfortable as an IC or architect, he focuses on pragmatic solutions that balance product needs, reliability and performance. Based in Tokyo, he pairs a PhD in speech and statistical learning with sustained open-source impact, including performance-driven contributions to widely used Python libraries.
19 years of coding experience
14 years of employment as a software developer
MSc, Signal processing and statistics, MSc, Signal processing and statistics at Ecole nationale supérieure des TélécommunicationsPhD, Computer Science, PhD, Computer Science at Kyoto University
English, French, Japanese
Stackoverflow
79,396reputation
10.0mreached
176answers
32questions
overloading
top-5%
shell
top-1%
audio
top-5%
matplotlib
top-5%
makefile
top-5%
plot
top-5%
Github Skills (37)
oop10
python10
capi10
c-programming10
attr10
subprocess10
attribute10
shell10
numerical-computing10
matplotlib9
float329
programming-language9
scipy9
algorithm9
refactor9
Programming languages (14)
Github contributions (5)
numpy/numpy
Nov 2007 - Nov 2014
The fundamental package for scientific computing with Python.
Role in this project:
Backend Developer
Backend DeveloperContributions:1664 commits, 17 comments, 1 issue in 7 years 1 month
Contributions summary:Cournapeau made several contributions to the NumPy library, focusing on C-level code modifications. They implemented new functions and enhanced existing functionalities within the core mathematical routines, primarily by adding tests for the `hypot`, `atan2`, and `copysign` functions as well as modifying the implementation of the correlate function. The user also addressed several bug fixes, addressing both performance improvements and the handling of special values such as NaN and infinity in existing functions.
lapackpythonmpindarrayconvolution
python-attrs/attrs
Oct 2015 - May 2016
Python Classes Without Boilerplate
Role in this project:Back-end Developer
Back-end DeveloperContributions:7 commits, 4 PRs, 20 comments in 7 months
Contributions summary:Cournapeau focused on optimizing the `attrs` library's performance by refactoring and streamlining internal processes. They improved validation and conversion speeds by directly accessing private attributes and iterating over them more efficiently. Additionally, the user contributed to documentation and made minor improvements related to code maintainability.
pythonboilerplatepython3classes-pythonoop
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.