Xu Liang
NLP Researcher at Money Forward, Inc.
Tokyo, Japan
Summary
🤩
Rockstar🎓
Top SchoolXu Liang is an NLP researcher and engineer with 11 years of experience building production-ready language solutions, currently developing RAG-based recommendation systems at Money Forward in Tokyo. He combines hands-on model building (ML/DL), prompt engineering, and practical LLM research with a strong track record in information extraction, relation extraction, and RDF/knowledge-graph construction. A prolific OSS author, he has published tools like JCLdic, SeqAL, and SegAL and contributed front-end improvements to the widely used doccano annotation tool. Fluent in Chinese, Japanese, and English, he pairs technical depth with data storytelling and visualization skills to turn complex analyses into actionable reports. Notably, his work on applying OpenAI models to new business workflows led to validated RAG prototypes and patentable product ideas.
11 years of coding experience
6 years of employment as a software developer
大连大学Master's degree, Neural Network, Master's degree, Neural Network at 日本早稻田大学
日语（日本語能力試験n1レベル）, 英语（toeic 940，toefl 97）, Chinese
Stackoverflow
331reputation
18kreached
2answers
5questions
Github Skills (25)
javascript10
python10
css10
pandas10
time-series10
vuejs310
front-end-development10
data-cleaning10
vuejs210
vue10
vue-jsx10
data-analysis10
mysql9
annotations9
redis9
Programming languages (10)
Github contributions (5)
BrambleXu/pydata-notebook
Oct 2017 - May 2018
利用Python进行数据分析 第二版 (2017) 中文翻译笔记
Role in this project:
Data Scientist
Data ScientistContributions:166 commits, 14 PRs, 85 pushes in 6 months
Contributions summary:Xu's commits focus on analyzing and interpreting the provided data using Python. The contributions involve reading, processing, and manipulating data from the USA.gov dataset, with an emphasis on understanding time-series data and identifying the key features present within the dataset. The user is involved in exploring the data, with the objective of uncovering patterns and providing actionable insights. The code demonstrates skills in data cleaning, data transformation, and time series analysis.
data-analysispythonpandas
Show-Me-the-Code/python
Oct 2016 - Nov 2016
Show Me the Code Python version.
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:13 commits, 10 PRs in 1 month
Contributions summary:Xu primarily focused on solving introductory programming problems, showing a proficiency in basic Python concepts. They implemented solutions for a few simple coding challenges, including generating activation codes and resizing images. Further contributions included integrating and experimenting with database technologies such as MySQL and Redis to store generated data.
python-codepythonpython3python-version
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.