Hello, I am

Jangseop Park

Ph.D. Candidate, Smart Design Lab · CCS Graduate School of Mobility, KAIST

I explore the intersection of Large Language Models and Mechanical Engineering,
developing Agentic AI for autonomous engineering design and 3D Surrogate Modeling for design optimization.

About Me Contact

About Me

A brief introduction

Jangseop Park

Ph.D. Candidate, KAIST

Jangseop Park (박장섭) is a Ph.D. Candidate at the Smart Design Lab, KAIST, advised by Prof. Namwoo Kang.
He previously worked as an AI Researcher at NARNIA LABS, specializing in generative AI for industrial product development.

His research explores the intersection of Large Language Models and Mechanical Engineering,
developing Agentic AI for autonomous engineering design and 3D Surrogate Modeling for design optimization.

Keywords: Agentic AI for Engineering Design · 3D Surrogate Modeling for Design Optimization

Download CV

Background

Education and experience

Education

Ph.D. Candidate, CCS Graduate School of Mobility
Mar. 2023 – Feb. 2027

KAIST · Advisor: Prof. Namwoo Kang

M.S., CCS Graduate School of Mobility
Aug. 2020 – Feb. 2023

KAIST · Advisor: Prof. Namwoo Kang

B.S., Mechatronics Engineering
Feb. 2012 – Aug. 2019

KOREATECH · Advisor: Prof. Sangsoon Lee

Experience

Graduate Researcher
Sep. 2020 – Present

Smart Design Lab, KAIST

AI Researcher
Feb. 2023 – Aug. 2024

Narnia Labs

Mechanical Design Engineer
May. 2019 – Oct. 2019

ATI — Advanced Technology Inc.

Student Researcher
Apr. 2017 – Dec. 2018

Computer-Aided-Engineering Lab., KOREATECH

Highlights

Proposed frameworks and recent publications

Proposed Frameworks

Agentic AI
AutoATC

An LLM-driven framework that automatically decomposes complex multidisciplinary design problems.

3D Surrogate Modeling
Point-DeepONet

Predicts nonlinear physical fields on arbitrary 3D shapes under varying load conditions.

Selected Publications

  1. AutoATC: Automated Analytical Target Cascading Decomposition with LLM-Based Multi-Agent System.

    J. Park, N. Kang*.

    Under review, 2026.

  2. Point-DeepONet: Predicting Nonlinear Fields on Non-Parametric Geometries under Variable Load Conditions.

    J. Park, N. Kang*.

    Neural Networks, 108560, 2026.

    PDF Code

Recognition

Selected awards and teaching

Awards
  • Best Presentation Award — KSME ConferenceMay. 2022
  • Best Presentation Award — KSCE ConferenceNov. 2021
  • Minister's Award — Science & Technology, Capstone Design FairOct. 2018
Teaching Assistant