A hands-on builder who takes on any opportunity to learn through experience.
I enjoy exploring unfamiliar fields, turning ideas into working projects, and learning from real implementation challenges.
- Interested in data analysis and discovering actionable insights from real-world datasets
- Exploring indoor spatial reconstruction using depth estimation and camera pose analysis
- Learning reliable RAG pipelines, retrieval quality, and practical LLM integration
- Interested in infrastructure architecture, cloud deployment, and scalable system operations
| Project | What I Built | Core Technologies |
|---|---|---|
| Musubi Movie Recommendation Site | A full-stack movie discovery platform with personalized recommendations and RAG-powered character conversations | React, FastAPI, PostgreSQL, Milvus, Docker, Kubernetes |
| Spatial Floorplan Lab | An experimental pipeline that converts handheld indoor RGB video into rough 2D floor-plan drafts | Python, COLMAP, YOLO, Depth Estimation, OpenCV |
| AI Job Market Trend Analysis | A reproducible classification experiment for high-salary AI job analysis with model comparison and experiment tracking | pandas, scikit-learn, XGBoost, LightGBM, MLflow, DVC |
