Chris Chen （陈庆宇）

Hello! I'm Chris Chen, a master's student in Electrical and Computer Engineering at the University of Washington.

I'm interested in AI agents, model-serving systems, and backend engineering. I like learning new technologies and understanding how they work by building with them. I hope that, one day, I can create technology that is both useful and genuinely interesting.

I came to computer science through an unconventional path. I studied education as an undergraduate, then taught myself mathematics and computer science and transitioned into software engineering. That experience still shapes how I approach unfamiliar problems: start from the foundations and keep working until the system makes sense.

GitHub

Selected projects

  1. InferOps Agent. An LLM inference optimization agent that plans, executes, and evaluates vLLM experiments with retrieved knowledge and experiment history.

    Python, LangGraph, vLLM, SQLite, MLflow · 153 tests, 5 workloads

  2. C++ High-Concurrency URL Shortener. A RESTful short-link service built with Linux epoll and a thread pool, including Redis cache protection, MySQL sharding, Kafka events, and Prometheus metrics.

    C++, Linux, Redis, MySQL, Kafka, Prometheus

  3. Recommendation Serving and Observability. A reproducible model-serving prototype connecting recommendation APIs, runtime model loading, caching, load testing, and observability.

    FastAPI, Redis, MLflow, Docker, Prometheus, Grafana

Research and experience

  1. Multimodal SWE-bench Data Curation. Built a GitHub API filtering pipeline and reviewed more than 300 issues to curate reproducible multimodal software-engineering tasks.

    University of Washington Tech Policy Lab · 2026–present

  2. OS-Marathon. Curated about 20 academic-document workflows and studied agent failures in visual grounding, state tracking, and repetitive interaction.

    ACL Rolling Review submission · Fifth author · 2026

  3. Machine Learning Engineer Intern. Developed HD-map validation tools and visualization tools for BEV features from a six-camera semantic mapping model.

    Changan Automobile, Advanced Technology Research Institute · 2023

Education

Technical interests

AI Agents, MCP, RAG, Agent Evaluation; vLLM, FastAPI, MLflow; Linux, Redis, MySQL, Kafka, Docker, Prometheus; Python, C/C++, SQL.

Outside of work

I enjoy sports, books, films, and watching football.