GitHub
Prognostics and Health Management
Fault diagnosis, anomaly detection, and predictive maintenance for rotating machines, thermal power plants, and complex industrial systems.
Personal Homepage
Hyeongmin Kim
Hi, I am Hyeongmin. I am currently a Staff Engineer at Samsung Research, and this is my small home for work, research notes, and the things I have been building around engineering AI.
I started from mechanical engineering, spent my graduate and postdoctoral years studying PHM and domain-informed AI at Seoul National University, and now work on practical AI systems that use knowledge graphs, LLMs, and RAG.
About
A quick intro.
This site is mainly a personal introduction page. It keeps together my background, research interests, old study notes, and links that are useful when someone wants to know what I work on.
I received my B.S. in Mechanical Engineering from Seoul National University in 2019, then continued Ph.D. and postdoctoral research at HAI LAB. My work has mostly lived between mechanical systems and AI: fault diagnosis, domain-informed neural networks, design optimization, and LLM-based engineering workflows.
PHM
Domain-informed ML
Engineering Design
Knowledge Graphs
LLM/RAG Agents
Research
What I Work On
Domain Knowledge Informed Neural Networks
Neural networks that use physics, signal analysis, expert rules, or structured domain priors to improve generalization and interpretability.
AI-driven Design
AI methods for design generation, performance prediction, inverse modeling, reinforcement learning, and design parameter optimization.
LLM Applications in Engineering
Engineering datasets, knowledge graphs, RAG workflows, and LLM agents that can support design reasoning, signal analysis, and physical validity checking.
Research Thread
Making AI less black-box for engineering problems.
A recurring theme in my work is that engineering AI should not stop at numerical prediction. It should expose useful structure: what signal patterns matter, which physical constraints are being respected, and how a human engineer can use the result.
Experience
Path
Staff Engineer, Samsung Research
Working on practical AI systems that connect domain knowledge, LLMs, RAG, and intelligent agents for engineering applications.
HAI LAB, Seoul National University
Research on PHM, DINN, AI-driven design, and engineering applications of domain-informed machine learning.
B.S. Mechanical Engineering, Seoul National University
Foundation in mechanical systems, design, dynamics, and engineering analysis.
Activities
Selected Presentations and Visits
May 2024 - Zhengzhou, China
ACSMO 2024
Window Size and Sampling Rate Selection for Cost-optimal Deep Learning-based Fault Diagnosis.
August 2023 - Boston, USA
ASME-IDETC 2023
Convolutional Auto-Encoder-Based Boiler Tube Leakage Detection Method in a Thermal Power Plant.
August 2022 - St. Louis, USA
ASME-IDETC 2022
A Multi-scale Residual Network with Attention Mechanism for Fault Diagnosis of Rotating Machines.
June 2022 - August 2022 - Aachen, Germany
Visiting Scholar, Fraunhofer IPT
Collaborative research on inverse modeling and reinforcement learning for design parameter optimization in manufacturing.
Extra
Older notes and study logs.
I keep older lecture notes and study materials in a separate archive so this homepage can stay focused while the earlier material remains available.
Open extra archiveContact