Kevin Hu
Senior Algorithm Engineer Algorithm Team Manager at YICHENG Technology co., ltd.
Shanghai, Shanghai, China
Summary
🤩
Rockstar🎓
Top SchoolKevin Hu is a Senior Algorithm Engineer and team manager in Shanghai with 17+ years designing search, recommendation and NLP-driven systems across finance, e-commerce and consumer apps. He has led end-to-end projects from knowledge-base QA and taxonomy generation to large-scale vector search and production RAG pipelines, combining C/C++, Python, Spark, Hive and ElasticSearch/Solr in high-throughput environments. At YICHENG he applied BERT-derived embeddings for DSSM, personalized query suggestion and hybrid re-ranking using features like embedding similarity and XGBoost, while earlier roles included CTO-level responsibility for search and messaging architectures. An early contributor to the open-source RAGFlow project, he implemented robust PDF parsing, document understanding and Dockerized deployment components, reflecting a practical focus on making deep document understanding production-ready. He pairs deep algorithmic experience with hands-on infrastructure work and a proven track record of boosting business metrics through principled ML and engineering.
17 years of coding experience
11 years of employment as a software developer
Master degree; B.S, Computer Software & Theory; Computer Science; Human face detection;, Master degree; B.S, Computer Software & Theory; Computer Science; Human face detection; at Central South University
Github Skills (12)
retrieval-augmented-generation10
pdf-processing10
docker10
parsing10
python10
parse10
dockers10
retrieval-augmented10
document-processing10
postgresql9
nlp8
machine-learning8
Programming languages (4)
Github contributions (5)
infiniflow/ragflow
Apr 2024 - Apr 2025
RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:6 releases, 275 reviews, 2821 PRs in 1 year
Contributions summary:Kevin implemented the initial Python code, specifically focusing on PDF parsing with `pdfplumber`, highlighting expertise in document understanding and processing within the RAGFlow framework. Further contributions include setting up PostgreSQL database connections and implementing data structures, as well as incorporating the `huqie` library for text analysis. The user also added Docker Compose files, adding to system's infrastructure and ease of deployment, and integrating image-to-text functionality.
agentagentsai-searchchatbotchatgpt
KevinHuSh/ragflow
Apr 2024 - Apr 2025
RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.
Contributions:1737 pushes, 12 branches in 1 year
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.