CV

This is a description of the page. You can modify it in '_pages/cv.md'. You can also change or remove the top pdf download button.

Contact Information

Name Qi Chen
Professional Title Applied Scientist
Email 7qchen@gmail.com
Location San Francisco Bay Area, California

Professional Summary

Applied Scientist at AWS working on agentic AI for software engineering, including coding agents, evaluation, and benchmarks.

Experience

  • 2024 -

    San Francisco Bay Area, CA

    Applied Scientist II - Agentic AI for Software Engineering
    Amazon Web Services (AWS)
    • Designed VibeCodingBench to benchmark coding agents in real-world scenarios with simulated human collaboration.
    • Built a dual evaluator for repository-level code documentation agents, resulting in a first-author U.S. patent.
    • Initiator and principal contributor for /doc, a repository-level documentation agent in Amazon Q Developer, launched at AWS re:Invent 2024 as a keynote demo.
    • Implemented CodeAct-style scaffolding in Strands Agents for agent self-improvement via ‘think in code’.
    • Exploring memory-layer designs for coding agents to support user personalization and continual learning.
  • 2023 - 2023

    Beijing, China

    Machine Learning Engineer - Kwai Star Talent Program
    Kuaishou Technology (Kwai)
    • Core algorithm designer for CREAD, a company-wide watch-time prediction framework in Kwai’s video recommendation system.
    • Achieved 5.85% improvement in XAUC and 0.29% lift in user watch time in A/B testing.
    • CREAD deployed across multiple major product lines serving 50M+ daily active users.
    • Work published at AAAI 2024. Cited by multiple papers including the ACM RecSys 2025 best paper.

Education

  • 2017 - 2023

    Beijing, China

    Ph.D.
    Peking University
    Applied Mathematics
    • Thesis: Improving Deep Equilibrium Models with Global Information
    • Advisors: Prof. Zhouchen Lin, Prof. Yisen Wang, Prof. Jiansheng Yang
  • 2013 - 2017

    Beijing, China

    B.S.
    Peking University
    School of Mathematical Sciences

Awards

  • 2025
    Top Reviewer of NeurIPS
    NeurIPS
  • 2022
    Ubiquant Scholarship
    Peking University

    Top 3%

  • 2017
    Dean's Scholarship
    Peking University

    Top 25%

Skills

Agentic AI (Expert): Coding Agents, LLM Agents, Benchmarks, Evaluation
Machine Learning (Expert): Deep Equilibrium Models, Graph Neural Networks, Recommendation Systems
Programming (Expert): Python, PyTorch, AWS

Languages

Chinese : Native speaker
English : Fluent

Interests

AI for Software Engineering: Coding Agents, Long-horizon Coding Tasks, Benchmarks, LLMs
Deep Learning Theory: Deep Equilibrium Models, Graph Learning, Implicit Neural Networks