CV

Contact Information

Name Han Zhang
Professional Title PhD Student in Computer Science
Email hhzhang@cs.ucla.edu

Professional Summary

PhD student at UCLA ScAI Lab, interested in Agentic RL, Agentic AI systems, and reinforcement learning-based fine-tuning methods.

Experience

  • 2025 - 2025

    Remote

    Data and Research Intern
    Optum AI
    • Mentor: Kimmo Kärkkäinen
    • Built a two-stage LLM finetuning pipeline for ICD-10-CM diagnosis prediction
    • Developed hierarchy-aware contrastive learning and dynamic confidence loss for ICD code classification
    • Scaled dataset from 100K to 2.5M+ samples using PySpark
  • 2024 - 2024

    Shanghai, China

    Applied Scientist Intern
    Amazon Shanghai AI Lab
    • Mentors: Dr. Minjie Wang, Quan Gan
    • Proposed GFS and integrated into Project 4DBInfer
    • Explored novel strategies for restructuring relational databases before graph construction
  • 2022 - 2023

    Shanghai, China

    Undergraduate Researcher
    APEX Lab, Shanghai Jiao Tong University
    • Advisor: Prof. Weinan Zhang, Dr. David Wipf
    • Led the GFS project on deep learning over relational databases
    • Developed feature space alignment for tabular data using hard negative contrastive pretraining
  • 2023 - 2024

    Pittsburgh, PA, USA

    Research Intern
    Ma Lab, Carnegie Mellon University
    • Advisor: Prof. Jian Ma
    • Applied CNNs and transformer-based models to computational genomics
    • Developed novel attention layers for Hi-C data to predict gene expression

Education

  • 2024 - present

    Los Angeles, CA, USA

    PhD
    University of California, Los Angeles
    Computer Science
    • Advisor: Prof. Wei Wang
    • GPA: 4.0/4.0
  • 2020 - 2024

    Shanghai, China

    Bachelor of Engineering
    Shanghai Jiao Tong University
    Computer Science (ACM Honor Class)
    • Major GPA: 91.87/100 (Rank 5/35)

Awards

  • 2023
    Ruiyuan-Hongshan Scholarship
    ACM Class, SJTU

    Top 5% students in ACM class

  • 2021, 2022, 2023
    Zhiyuan Honorary Scholarship
    SJTU

    Top 5% students in SJTU

  • 2021
    Jin Long Yu Scholarship
    SJTU

Skills

Programming (): Python, C/C++, Java, Rust, Verilog
Research (): Agentic RL, LLM Post-Training, LLM Agents

Languages

Chinese : Native speaker
English : Fluent