👋 Hi there, I’m Jinhee!

M.S. in Artificial Intelligence student at UTSA, based in San Antonio. Drawing on my experience in the banking industry, I’m interested in building reliable and impactful AI solutions for financial data.

📄 View Resume

---

🎓 Education

• M.S. in Artificial Intelligence — UT San Antonio (In Progress)
• B.S. in Computer Science — Korea National Open University (2023–2025)
• B.A. in Police Administration & Chinese Language — Soonchunhyang University (2014–2020)

💼 Experience

• Software Developer at KB Kookmin Bank, South Korea (2023–2025)
• Retail Banking Associate at KB Kookmin Bank, South Korea (2020–2022)

🛠 Tech Stack

🔭 Research Interests

• Explainable AI — SHAP, LIME, DiCE
• LLMs & RAG — Retrieval-Augmented Generation
• Computer Vision — Object Detection (Faster R-CNN, YOLO)

🧱 Featured Projects

Finance RAG Chatbot

A RAG chatbot that retrieves and summarizes information from complex financial PDFs.

• Custom semantic chunking based on financial clause structure (regex)
• MMR + LLM-based reranking (Upstage Solar Pro 2) to improve retrieval accuracy
• Interactive web UI built with Streamlit

Stack: Python 3.12 · LangChain · ChromaDB · Streamlit · Upstage Solar API View on GitHub →

---

Explainable Credit Default Prediction (XAI)

Comparative study on credit card default prediction using SHAP and DiCE for model interpretability.

• SHAP for feature attribution (Logistic Regression) vs. DiCE for counterfactuals (XGBoost)
• Generated “what-if” scenarios showing minimal changes needed to flip a prediction

Stack: Python · Scikit-learn · XGBoost · SHAP · DiCE · Pandas View on GitHub →