John Selvam

Senior Data Scientist

United States
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

👤
Senior
🎓
Top School
John Selvam is a Senior Data Scientist with nine years of experience turning academic ML research into production-ready systems for companies like Walmart Labs and Intuit. He specializes in feature engineering, time-series temporal disaggregation, and credit scorecard modeling—publishing useful open-source packages (monotonic-binning and timedisagg) that have attracted community adoption. His background blends an electrical engineering degree from NIT Calicut and a Masters in Analytics from the University of Minnesota, giving him both theoretical depth and practical rigor. At Walmart he focused on productizing cutting-edge models; at Equifax he built customer segmentation, churn, and PD scorecard solutions that supported revenue and retention decisions. He prefers to let code speak for him—most of his signal is in his GitHub repos where he publishes niche, reusable ML tools. Reachable by email, he actively seeks collaboration and contributions that move research into reliable production.
code9 years of coding experience
job5 years of employment as a software developer
bookMasters in Analytics, Masters in Analytics at University of Minnesota
bookBachelor's degree, Electrical Engineering, Bachelor's degree, Electrical Engineering at National Institute of Technology Calicut
github-logo-circle

Github Skills (20)

binning10
pip10
python9
statistics9
forecast8
data-analysis8
forecasting8
machine-learning8
autoencoder8
r8
data-science8
adversarial-learning7
scikit-learn7
sequential6
variational-autoencoder6

Programming languages (1)

Python

Github contributions (5)

github-logo-circle
jstephenj14/timedisagg

May 2020 - Dec 2020

Python package that implements temporal disaggregation models to convert low-frequency to high-frequency time series (pip install timedisagg).
Contributions:20 commits, 8 PRs, 10 pushes in 7 months
pippythontime-series
Python package that optimizes information value, weight-of-evidence monotonicity and representativeness of features for credit scorecard models (pip install monotonic-binning)
Contributions:50 commits, 15 PRs, 46 pushes in 4 years 8 months
binningpippython
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