Zhexiong Liu
Data & Applied Scientist II
Pittsburgh, Pennsylvania, United States
Summary
👤
Senior🎓
Top SchoolZhexiong Liu is a Data & Applied Scientist II at Microsoft with a Ph.D. in Computer Science from the University of Pittsburgh and eight years of experience bridging NLP research and production ML. He has a strong track record of internships and research roles at Microsoft, Amazon, Emory NLP, and Pittsburgh, focusing on robust language transformations and applied modeling. His open-source contributions include enhancing the widely used GEM-benchmark NL-Augmenter—developing an antonym substitution transformation that handles POS tagging, WordNet antonyms, and tricky edge cases to improve data augmentation reliability. Zhexiong combines academic rigor (cross-registration coursework at Carnegie Mellon) with hands-on engineering, moving ideas from papers into production during multiple Microsoft internships. Based in Pittsburgh and known for a quirky enthusiasm for cats, he brings practical solutions to NLP robustness and data-centric ML challenges.
8 years of coding experience
5 years of employment as a software developer
Bachelor of Engineering Computer Science and Technology, Bachelor of Engineering Computer Science and Technology at Sichuan UniversityMaster of Science Computer Science, Master of Science Computer Science at Emory UniversityDoctor of Philosophy Computer Science, Doctor of Philosophy Computer Science at University of PittsburghCross Registration Causal Inference & NLP, Cross Registration Causal Inference & NLP at Carnegie Mellon University
Chinese, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (8)
nltk10
nlp10
spacy10
wordnet10
python10
machine-learning9
text-generation9
tokenizer8
Programming languages (3)
Github contributions (5)
GEM-benchmark/NL-Augmenter
Aug 2021 - Oct 2021
NL-Augmenter 🦎 → 🐍 A Collaborative Repository of Natural Language Transformations
Role in this project:
ML Engineer
ML EngineerContributions:6 reviews, 18 commits, 2 PRs in 1 month
Contributions summary:Zhexiong primarily contributed to the `antonyms_substitute` transformation within the NL-Augmenter repository. Their work involved developing code to substitute words with their antonyms using spaCy for POS tagging and nltk for wordnet-based antonym retrieval. The user refined the transformation logic by addressing edge cases with synonyms/antonyms, and adjusting the code to ensure only even number of substitutions happen. Furthermore, the user added keywords to enhance discoverability.
nlpinformation-theorysentencedeep-learningmachine-learning
ZhexiongLiu/Nasdaq-ITCH-5.0
Mar 2019 - Sep 2019
Contributions:13 commits, 11 pushes, 1 branch in 6 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.