Hongpeng Guo is a Senior Research Scientist at ByteDance, specializing in large-scale machine learning systems. With a Ph. D. in Computer Science from the University of Illinois Urbana-Champaign, his expertise lies in developing infrastructure for reinforcement learning and foundation model training. His background includes roles at Anyscale, Google, and Jane Street.
He developed "Verl, " an open-source framework designed to make reinforcement learning for large language models more efficient and scalable.
Hongpeng Guo
Senior Research Scientist at ByteDance
San Francisco, California, United States
Overview
Topics They Care About
Scalable Reinforcement Learning
He develops large-scale RL infrastructures at ByteDance and co-created Verl, an open-source framework for building end-to-end RL pipelines with large language models.
Foundation Model Training
His team at ByteDance focuses on building impactful systems for both pre-training and post-training of large language and vision models.
Distributed ML Systems
His experience includes building Ray Train at Anyscale and working on Google Cloud's core infrastructure, focusing on systems for large-scale ML workloads.
Media Appearances
Hongpeng has no verified media appearances
Work History
3-2025
Senior Research Scientist at ByteDance
4-2024 - 3-2025
Machine Learning Engineer at Anyscale
5-2023 - 8-2023
Quantitative Trading Intern at Jane Street
5-2022 - 8-2022
Software Engineer Intern (Machine Learning) at Meta
5-2021 - 8-2021
Software Engineer Intern at Google
Education
2018 - 2-2024
Doctor of Philosophy - PhD from University of Illinois Urbana-Champaign
2014 - 2018
Bachelor of Engineering - BE from The University of Hong Kong
More Information
Social Presence :
Prographics :
Exp : 2
Location : San Francisco, California, United States
Job Level : N/A
Designation : Senior Research Scientist at ByteDance
