Xingyu Wang.

Resume

Xingyu Wang

AI / LLM Engineer building reliable, production-scale intelligent systems.

Get in touch

Profile

I design and build production AI systems—from multi-agent workflows and retrieval pipelines to the infrastructure, observability, and performance work that keeps them dependable. My background spans data engineering, machine learning, and full-stack development.

70%
agent workflows migrated
~30K
daily users supported
10–20%
performance improvement

Experience

AI / LLM Engineer

Production AI · Judicial technology

  • Led the migration of approximately 70% of judicial agent workflows to LangGraph, improving maintainability and performance by 10–20%.
  • Designed multi-agent orchestration, state management, observability, and RAG pipelines for a platform serving approximately 30,000 daily users.
  • Conducted load testing and production optimization across LLM inference, retrieval, and agent workflows.

Data Engineer

Machine learning · LLM-powered data products

  • Re-architected an NFT valuation system, reducing the codebase from more than 20K lines to approximately 4K and introducing LightGBM into the modeling pipeline.

Data Analyst

Environmental Models · Government systems

  • Developed environmental mathematical models and containerized data workflows with Docker.
  • Contributed to data modeling and system delivery for government environmental information platforms.

Education

Doshisha University

Master of Engineering · Mathematical Environmental Science

Wuhan University

Bachelor of Science · Geography

Capabilities

AI systems
LLM applications · Multi-agent systems · LangGraph · RAG
Engineering
Python · TypeScript · SQL · Django · React
Data
Data engineering · Mathematical modeling · Machine learning