Josef Perktold
Montreal, Quebec, Canada
Summary
🤩
Rockstar🎓
Top SchoolJosef Perktold is a Python-focused statistician and software engineer with 17 years of experience who has shifted from economics to improving statistical tooling for the Python ecosystem. He has contributed significant test coverage, bug fixes, and methodological enhancements to SciPy and is actively building out statsmodels into a robust statistics and econometrics package. His work emphasizes reliability and verified testing for statistical distributions and rates models, including new tests for multivariate covariance structures and practical functions like TOST. Based in Montreal, he blends domain expertise in econometrics with disciplined QA and backend development practices. Notably, his contributions to the widely used SciPy library have helped raise precision and correctness in core statistical routines relied on across science and engineering. He communicates and coordinates improvements openly through project mailing lists, reflecting a collaborative open-source approach.
17 years of coding experience
University of California, Los Angeles
German, English
Stackoverflow
22,587reputation
2.6mreached
418answers
3questions
matrix
top-5%
numpy
top-1%
machine-learning
top-5%
regression
top-1%
python
top-1%
unit-testing
top-5%
Github Skills (27)
scipy10
python10
testing10
statistics10
hypothesis-testing10
statistic10
numpy10
regression10
statistical-models10
hypothesis-tests10
scientific-computing10
data-analysis10
pandas9
unit-testing9
matrix9
Programming languages (6)
Github contributions (5)
statsmodels/statsmodels
Dec 2008 - Jan 2023
Statsmodels: statistical modeling and econometrics in Python
Role in this project:
Data Scientist
Data ScientistContributions:5 releases, 197 reviews, 3818 commits in 14 years 3 months
Contributions summary:Josef implemented a new Tost function and performed code cleanup within the statsmodels/stats/rates.py file. They also added the method='etest' functionality to the test_poisson_2indep function and applied changes to test functions for rates Poisson models. Additionally, the user contributed to multivariate analysis tests by adding a new test for covariance structure. These changes indicate a focus on enhancing the statistical analysis capabilities of the repository.
forecastingpythonregression-modelsstatsmodelsstatistics
scipy/scipy
Nov 2008 - Aug 2013
SciPy library main repository
Role in this project:Back-end Developer & QA Engineer / Test Automation Engineer
Back-end Developer & QA Engineer / Test Automation EngineerContributions:17 reviews, 159 commits, 1123 comments in 4 years 9 months
Contributions summary:Josef primarily contributed to the development of statistical methods within the SciPy library. The commits demonstrate the implementation and testing of statistical distributions, including adding tests for existing methods. They also worked on correcting existing functions related to these statistical methods and fixing issues related to parameter handling and precision. The user's work improved the reliability and accuracy of several statistical functions within the library.
scipypythonscientific-computing
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.