Usman Khan
Machine Learning Research Scientist at Meta
Menlo Park, California, United States
Summary
👤
Senior🎓
Top SchoolUsman Khan is a Machine Learning Research Scientist at Meta with a PhD from North Carolina State University and a decade of experience building ML systems that span signal processing, radar sensing, and recommendation ranking. His work blends academic rigor—publishing at venues like WWW and UbiComp—with production-focused ML: from feature engineering and online A/B experiments at Meta to deep learning pipelines for radar-based localization and health monitoring at Nokia Bell Labs and Koko. He has a strong open-source streak, contributing backend I/O, testing, and simulation features to the Stingray time-series library used in astronomical analyses. Comfortable moving models from data collection and preprocessing through training to deployment, he brings both experimental inventiveness and pragmatic engineering discipline. Based in Menlo Park, he combines electrical engineering roots with modern ML research to tackle perceptual and large-scale ranking problems.
10 years of coding experience
4 years of employment as a software developer
PhD, Computer Science, 4.0, PhD, Computer Science, 4.0 at North Carolina State UniversityElectrical Engineering, Electrical Engineering at NUST SEECS
Urdu, German, English
Github Skills (11)
time-series10
python10
astrophysics10
astronomy10
pickle10
data-analysis10
frequency-analysis9
pytest9
f9
fourier-transform9
hdf8
Programming languages (6)
Github contributions (5)
StingraySoftware/stingray
Mar 2016 - Mar 2017
Anything can happen in the next half hour (including spectral timing made easy)!
Role in this project:
Back-end Developer
Back-end DeveloperContributions:132 commits, 20 PRs, 99 comments in 1 year
Contributions summary:Usman's contributions center around the development and refactoring of input/output methods for the "stingray" Python library, which appears to be used for astronomical time series analysis. Their work includes implementing methods to save and retrieve class objects using the pickle format. The user modified these functions, corrected data management, added file format read/write functionality and test suits for the aforementioned processes. Furthermore, the developer extended the library by adding energy channels for data simulations.
x-rayblackholedata-analysistime-series-analysispython
usmanwardag/usmanwardag.github.io
May 2016 - Apr 2020
About me.
Contributions:4 PRs, 60 pushes, 4 branches in 3 years 11 months
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.