Hanhui Su

Computer Science master's student and AI engineer.

I am currently pursuing an M.S. in Computer Science at UC Riverside. My background spans AI/ML engineering, LLM data work, embedded software, and product-minded engineering across research and industry settings.

About

A short introduction

I am Hanhui Su, a Computer Science master's student at UC Riverside. Before graduate school, I studied Internet of Things Engineering in Beijing and ranked in the top 5% of my department.

My path has moved between research and engineering environments: an LLM internship at the Institute of Software, Chinese Academy of Sciences, and an embedded software internship at Beijing Saishu Technology. Those experiences shaped the way I think about practical systems, reliability, and the gap between prototypes and usable products.

This website is meant to give a clear picture of who I am, how I have developed academically and professionally, and where my work is heading. Project details are separated below so the personal introduction stays focused.

Selected Work

Portfolio projects

Project 01

Evidence-Grounded RAG Agent

A financial document-analysis agent built with FastAPI, LangGraph, ChromaDB, and DuckDB for retrieval, tool execution, evidence planning, and grounded question answering over public company filings.

Open project

Project 02

Fin-Qwen

A reasoning-augmented distillation project that fine-tuned Qwen3-8B with QLoRA for financial social-media sentiment classification, using teacher-generated reasoning traces and structured JSON outputs.

Open project

Experience

Research and engineering background

May 2023 - Aug. 2023

LLM Intern, Institute of Software, Chinese Academy of Sciences

Processed 30 GB of unstructured Chinese eBook data for LLM pretraining, contributed to ArkPDFSFT data generation and validation, evaluated Llama 2 outputs, and ran LoRA fine-tuning experiments under GPU memory constraints.

Jul. 2021 - Sep. 2021

Embedded Software Engineer Intern, Beijing Saishu Technology

Contributed to a Bluetooth debugging tool, optimized networking through firmware updates, reduced device connection issues by 20%, and supported deployment of 50 IoT terminals across multiple sites.

Education

University of California, Riverside

M.S. in Computer Science, Sep. 2024 - Present

Beijing Information Science & Technology University

B.E. in Internet of Things Engineering, top 5% of the department

Skills

PyTorch Hugging Face LoRA / QLoRA Unsloth vLLM LangGraph ChromaDB DuckDB FastAPI Docker Python SQL