class ZhonghuiLi:
def __init__(self):
self.education = "M.S. in Computer Science & Engineering @ UC Santa Cruz"
self.role = ["Software Engineer", "Full Stack Developer", "AI/ML Engineer", "Data Scientist"]
self.languages = ["Python", "SQL", "Java", "JavaScript", "TypeScript", "C/C++"]
self.fun_fact = "I turn coffee into code and data into insights"
def say_hi(self):
print("Thanks for dropping by! Let's build something awesome together!")
me = ZhonghuiLi()
me.say_hi()Recent M.S. graduate in Computer Science and Engineering from UC Santa Cruz. Experience building full-stack applications and applied LLM systems, including ML-driven pipelines and production-ready services. Actively seeking full-time opportunities.
Open to: Software Engineer AI/ML Engineer Data Scientist Data Analyst Data Engineer Full Stack Backend
| Project | Description | Tech Stack |
|---|---|---|
| SkillMatch | Full-stack skill matching platform with ML recommendations (TF-IDF), real-time WebSocket messaging, and 85%+ test coverage | React, Express, Redis, Docker |
| 🍌 Slug Advisor · live demo | Deployed tool-calling agent for UCSC course advising — LangGraph ReAct, hybrid retrieval (BM25 + pgvector) + reranker, eval-in-CI gate, Langfuse observability + eval-score loop, MCP server | LangGraph, FastAPI, pgvector, Ragas, Langfuse, MCP, GCP Cloud Run |
| Autonomous Racing RL | Reinforcement learning agents (PPO) for autonomous vehicle racing simulation, with containerized HPC training pipeline | PyTorch, Docker, HPC, RL |
| Fine-Grained SSL | Cloud-scale image classification pipeline on GCP with semi-supervised training, achieving 93.84% accuracy | PyTorch, GCP, Semi-Supervised Learning |
| Meetily Fork | Extended Meetily with an MCP server connecting Claude Desktop to local SQLite meeting data for natural language session queries; includes automated startup scripting and planned LangGraph + BM25/FAISS agentic RAG pipeline | Python, FastAPI, Whisper, MCP, SQLite, Groq/Claude |
