Yunhai Hu is a computer science Masters student at NYU and a Research Collaborator at Yale, specializing in efficient Large Language Model inference. He has industry experience as a Machine Learning Engineer at bilibili and has co-authored multiple publications on AI model optimization.
He has co-authored multiple research papers on "Speculative Decoding, " a cutting-edge technique to accelerate the performance of large-scale autoregressive models.
Yunhai Hu
Research Collaborator at Yale University
Jersey City, New Jersey, United States
Overview
Topics They Care About
LLM Efficiency
His graduate research at NYU and publications, such as "PipeSpec, " are dedicated to making large language models faster and more efficient through new decoding techniques.
AI Systems
His background includes deep learning systems research and practical experience as a CUDA developer at bilibili, focusing on the infrastructure that powers AI.
Big Data Processing
Gained hands-on experience as a software developer at bilibili, working specifically as a big data engineer with technologies like Flink.
Media Appearances
Yunhai has no verified media appearances
Work History
12-2024 - 3-2025
Research Collaborator at Yale University
9-2024 - 2-2025
Research Assistant at New York University
9-2024 - 10-2024
Research Assistant at New York University
9-2023 - 9-2024
Machine Learning Engineer at bilibili Group
7-2022 - 9-2023
Software Developer at bilibili Group
Education
8-2024 - 5-2026
Master of Science - MS from New York University
9-2017 - 6-2022
Bachelor of Science - BS from Shandong University
More Information
Social Presence :
Prographics :
Exp : 2
Location : Jersey City, New Jersey, United States
Job Level : N/A
Designation : Research Collaborator at Yale University
