Youngseog Chung
Member Of Technical Staff at Microsoft AI
Pittsburgh, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top SchoolYoungseog Chung is a machine learning-focused software engineer with nine years of experience, currently a Member of Technical Staff at Microsoft AI after research and engineering roles at Google Brain, CMU Auton Lab, and an internship at Celonis. He holds advanced training in ML (PhD-level from Carnegie Mellon) and a quantitative CS/finance BS, combining rigorous research instincts with production-grade software delivery. His open-source contributions to the widely used uncertainty-toolbox demonstrate deep expertise in predictive uncertainty, implementing CRPS, Gaussian NLL, isotonic recalibration, and visualization features that improve model evaluation and calibration. A former Army sergeant, he brings disciplined systems thinking and a collaborative, results-driven mindset to multidisciplinary teams.
9 years of coding experience
2 years of employment as a software developer
Doctor of Philosophy - PhD Machine Learning, Doctor of Philosophy - PhD Machine Learning at Carnegie Mellon UniversityHong Kong University of Science and Technology (HKUST)
Github Skills (13)
scikit-learn10
uncertainty-quantification10
calibration10
python10
metric10
numpy10
scikit10
visualizations9
estimate9
visualization9
uncertainty9
bayesian-network9
matplotlib8
Programming languages (2)
Github contributions (5)
uncertainty-toolbox/uncertainty-toolbox
Oct 2020 - Sep 2021
Uncertainty Toolbox: a Python toolbox for predictive uncertainty quantification, calibration, metrics, and visualization
Role in this project:
Data Scientist
Data ScientistContributions:1 release, 10 reviews, 51 commits in 11 months
Contributions summary:Youngseog significantly contributed to the `uncertainty-toolbox` repository by implementing and integrating core metrics related to predictive uncertainty quantification. They added functionality for calculating the Continuous Ranked Probability Score (CRPS) and negative log-likelihood (NLL) for Gaussian distributions. Furthermore, the user introduced and refined recalibration methods for improving uncertainty estimates, including Isotonic Regression, and implemented plotting functionality for the visualization of uncertainty and recalibration results. Their work enhanced the toolbox's capabilities for evaluating and improving predictive models, as well as its visualization capabilities.
pythonpython-toolboxpredictive-uncertaintyvisualizationsrecalibration
YoungseogChung/NNKit
Mar 2020 - Apr 2020
Contributions:40 commits, 37 pushes, 1 branch in 24 days
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.