NLP Engineer · LLM Developer · Computational Linguist
I'm an NLP engineer and LLM developer with a graduate background in computational linguistics (NCCU; exchange at University of Tübingen). I work across the full NLP stack — from data pipelines and BERT fine-tuning to RAG systems and AI agent development.
My differentiator is combining linguistics with hands-on Natural Language Processing & LLM engineering — a rare mix that shapes how I design, evaluate, and interpret language models.
- 🏆 SemEval 2024: 1st in COPY accuracy, 2nd overall (numeral-aware generation)
- 📖 Master's thesis: GPT-4o classification on political discourse (F1: 0.80 → 0.89)
- 🎤 International conference presenter (TABU DAG held by the University of Groningen, the Netherlands)
- 🎓 MA in Linguistics — National Chengchi University (2021–2025)
- 🌍 Exchange, Computational Linguistics — University of Tübingen (2023–2024)
- 🎓 BMS in Information Management + BA in Foreign Language — National Cheng Kung University (2016–2021)
AI / NLP
Deployment & Tools
Core skills: LLM Fine-tuning · RAG · LangGraph · LlamaIndex · Prompt Engineering · Seq Labeling · NER
| Project | Description | Highlights |
|---|---|---|
| SemEval 2024 | Numeral-aware language generation | 🥇 1st COPY accuracy (82.17), 2nd overall |
| Appraisal Discourse Analysis | GPT-4o classification on PTT political comments + Streamlit dashboard | F1 0.80 → 0.89 via few-shot prompting |
| LegalTech Hackathon 2021 | AI-powered legal judgment analysis tool | Verdict probability prediction from court rulings |