AI Research Engineer & Data Communicator Bridging deep tech with real-world social insights.
I specialize in building autonomous data pipelines and multi-modal AI agents that understand human trends. My background spans Semiconductor Physics to Social Big Data Mining, giving me a unique "First Principles" approach to engineering.
- Currently working on: Multi-modal Agent Systems (converting images/video into structured market intelligence) & Local LLM Infrastructure (ROMA).
- Creator: Run a tech/data insight channel [์ํ๋ณํ๊ด์ธก์] with 120k+ subscribers, analyzing social phenomena through data.
Agentic AI & LLMs
Data Engineering
Cloud & DevOps
- WHOTAG (Global B2C Service): Built a Natural Language Search engine for influencers using LangGraph & Gemini.
- Tech: Google Cloud (Run, Vertex AI), Qdrant, Firestore.
- Role: Full-stack Data Engineering & AI Pipeline Architecture.
- Multi-modal Persona Analyzer: Automated system extracting "Gen Z Personas" from Instagram images/videos.
- Client: Samsung Global Research, K-Beauty Brands.
- Tech: HDBSCAN Clustering, UMAP, Multi-modal RAG.
- Text Mining for Mental Health (Frontiers in Psychiatry): Published paper on adolescent mental health during COVID-19 using social data mining.
- Trend Tracking System: Automated ETL pipelines processing TB-scale social data for clients like Riot Games, Hyundai, and LG Electronics.
- Paper: Korean adolescentsโ coping strategies... (Frontiers in Psychiatry)
- Book: 2024/2026 Trend Note (Co-author)
- Speaker: EBS Documentary (Low Birth Rate Special), Various Corp Seminars.
