Portrait of Pin Qian

Biography

My work focuses on the systems behind agents: reinforcement learning infrastructure, LLM inference, and production ML serving. I am currently a Machine Learning Engineer at Meta, working on Creative Ranking in Core Ads Growth.


Previously, I worked on reinforcement learning for game agents at Tencent, LLM inference research at Carnegie Mellon University, and reinforcement learning for portfolio optimization in earlier research. I am interested in building efficient and reliable systems that make agents practical in production.

Mail     GitHub     LinkedIn     Instagram     Blog

News


Education

Carnegie Mellon University Aug 2023 - May 2025

MS, AI Engineering, Electrical and Computer Engineering

University of Liverpool Sep 2019 - Jul 2023

BS, Computer Science

Industry Experience

Meta

Machine Learning Engineer, Creative Ranking, Core Ads Growth Feb 2025 - Present

  • Build format, content, cohort ranking models for large-scale ads delivery systems.
  • Optimize production ranking and delivery infrastructure, with a focus on serving efficiency, and end-to-end system scalability.

Software Engineer Intern, Machine Learning, Creative Delivery, Core Ads Growth May 2024 - Aug 2024

  • Worked on multimodal LLM-driven audience targeting in early-stage delivery systems.

Tencent Jun 2022 - Oct 2022

Machine Learning R&D Intern, Game AI Research Center

  • Worked on the PUBG game AI agent.
  • Trained RL-driven game agents and combined supervised learning to improve human-like and robust in-game behavior.

Research Experience

Infini-AI Lab, Carnegie Mellon University Jul 2024 - Dec 2024

Research Intern, ML Systems

  • Advised by Prof. Chen Beidi.
  • Conducted research on accelerating LLM inference using speculative decoding.

University of Liverpool Jan 2021 - Aug 2022

Research Intern, Reinforcement Learning

  • Conducted research on RL for portfolio optimization, with a focus on stochastic policies and robust sequential decision-making.
  • Co-authored a paper on interpretable stochastic model-free RL for portfolio optimization [paper].

Publications

Preview of the SPDQ paper

From deterministic to stochastic: an interpretable stochastic model-free reinforcement learning framework for portfolio optimization

Zitao Song, Yining Wang, Pin Qian, Sifan Song, Frans Coenen, Zhengyong Jiang, Jionglong Su.

Applied Intelligence, 2023. [paper]

Gallery

If a world model eventually emerges, I hope it will contain some of the beauty I have been lucky enough to see. Sony A7s3, 24-105/4, Zeiss 55/1.8, DJI Mini 3 Pro, DJI Mavic 4 Pro