Kaiyinzhou
Machine Learning Engineer at BUPT
Beijin Province, China
Summary
🤩
RockstarKaiyinzhou is an experienced Machine Learning Engineer based in Beijing with eight years applying deep learning and probabilistic methods to practical problems. He focuses on NLP and knowledge graphs, with hands-on expertise in variational inference and transformer-based models—evidenced by contributions to a BERT-based NER project where he implemented preprocessing, model training, and multi-class evaluation improvements. Comfortable moving models from research prototypes to robust training and evaluation pipelines, he balances algorithmic rigor with practical engineering. Colleagues would note his knack for squeezing extra performance out of established architectures and for clarifying complex probabilistic ideas in code.
8 years of coding experience
Github Skills (5)
tensorflow10
python10
bert10
machine-learning9
nlp9
Programming languages (2)
Github contributions (5)
kyzhouhzau/BERT-NER
Nov 2018 - Feb 2021
Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）.
Role in this project:
ML Engineer
ML EngineerContributions:60 commits, 1 PR, 55 pushes in 2 years 3 months
Contributions summary:Kaiyinzhou primarily contributed to a Named Entity Recognition (NER) project using Google's BERT model. Their work involved implementing core NER functionalities within the `BERT_NER.py` script, including data preprocessing, model definition, and evaluation metrics. They updated and refined the code related to model training, evaluation, and prediction, and also added a file for calculating multiple classifications. The changes suggest a focus on improving model performance and accuracy.
nlpnerentity-recognitionnamed-entity-recognitionbert
kyzhouhzau/Machine-Learning
Dec 2018 - Dec 2018
Contributions:8 commits, 2 PRs, 6 pushes in 3 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.