CV

AI Engineer

AIBIZ Global Co., Ltd.

• Improved anomaly detection by incorporating temporal dependencies into a GNN for semiconductor time-series data.
• Designed and developed a multi-level graph representation learning pipeline to capture both local and global sensor interactions.
• Implemented automated Root Cause Analysis using LangChain with DAPT.

Back End Developer

PFCT - PFC Technolgies

• Developed real-time stock data pipeline using Microservice Architecture.
• Designed and implemented RESTful APIs using Django and Django REST Framework.

PhD

North Carolina State University

Computer Science

MSc

Seoul National University of Science and Technology

Data Science

BSc

Seoul National University of Science and Technology

Information Technology Management

Anomaly Detection for Outgoing Quality Control of Camera Modules

LG Innotek, Inc.

• Extracted patterns of personal access behavior and labeled log data through statistical analysis.
• Developed an unsupervised LSTM model, decomposing log data into distinct keys and parameters.

ML-based Analysis of Cyclic Plasticity Behaviour of a Pipe Bend

LG Electronics, Inc.

• Enhanced the prediction accuracy of elastic shakedown boundaries through machine learning techniques.

A Self-Training based Machine Learning Platform for SOAR

AhnLab, Inc.

• Incrementally updated the tokenizer for newly generated web attack payloads.
• Implemented continual learning method for global model with local data to improve the performance.
• Developed REST APIs for data balancing, model training, and inference.

Realtime Prediction of Mechanical Equipment Failure based on Federated Learning

Kozen, Corp.

• Extracted patterns of personal access behavior and labeled log data through statistical analysis.
• Developed an unsupervised LSTM model, decomposing log data into distinct keys and parameters.

Anomaly Detection through Analysis of Large-Scale Personal Access Logs

Samo Convergence & Security, Corp.

• Extracted patterns of personal access behavior and labeled log data through statistical analysis.
• Developed an unsupervised LSTM model, decomposing log data into distinct keys and parameters.