Curtis Northcutt
Director, AI Research at Handshake
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolCurtis Northcutt is a Director of AI Research and serial founder with 11 years building practical, data-centric AI systems that scale from research labs to Fortune 500 deployments. He founded Cleanlab—creator of the widely used cleanlab Python package and confident learning framework—which was acquired by Handshake and is now the reliability layer that auto-fixes training data and attaches trust scores to model outputs. An MIT PhD whose work uncovered millions of label errors and earned the Morris Levins Thesis Award, Curtis has blended deep research (uncertainty quantification, noisy labels) with product-first engineering at Google, Oculus, Amazon, Facebook/FAIR and startups. He’s equally comfortable writing core package infrastructure (e.g., packaging and compatibility for cleanlab) as he is leading teams and shaping commercial AI reliability for enterprise customers. Notably, his research produced practical tools used inside Google and in academic benchmarks, revealing real-world dataset flaws that change how models are evaluated.
11 years of coding experience
13 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Computer Science, Doctor of Philosophy (Ph.D.) Computer Science at Massachusetts Institute of TechnologyCross Registration as MIT PhD Student Machine Learning, Cross Registration as MIT PhD Student Machine Learning at Harvard UniversityBachelor of Science Honors Computer Science and Mathematics, Bachelor of Science Honors Computer Science and Mathematics at Vanderbilt University
English, Spanish, c++, python, javascript (html, css), java, matlab, lisp, prolog, latex, vhdl, mips, sql
Stackoverflow
3,886reputation
5.2mreached
41answers
17questions
phantomjs
top-5%
scikit-learn
top-5%
validation
top-5%
python
top-5%
Github Skills (16)
packaging10
python10
package10
setuptools10
phantomjs9
scikit-learn9
validation9
confusion-matrix6
git6
casperjs6
logistic-regression6
roc6
machine-learning6
pipeline6
cross-validation6
Programming languages (7)
Github contributions (5)
cleanlab/cleanlab
May 2018 - Jan 2023
The standard data-centric AI package for data quality and machine learning with messy, real-world data and labels.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:6 releases, 210 reviews, 741 commits in 4 years 8 months
Contributions summary:Curtis's contributions primarily involve creating and updating core infrastructure files. Specifically, the user created a setup.py file, defining the package's name, version, license, and dependencies. They also implemented a version.py file. Furthermore, the user contributed to adding and updating files such as the license and python 3.7 compatibility to support the cleanlab package.
operationdata-preprocessingtext-annotationreal-world-datadata-labeling
cgnorthcutt/dotfiles
Feb 2019 - Sep 2022
Contributions:70 pushes, 1 branch in 3 years 7 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.