Research

Connecting large language models with mechanical engineering
to build autonomous design systems.

Research Interests

Agentic AI for Engineering Design

LLM-based multi-agent systems that reason about, decompose, and autonomously solve multidisciplinary engineering design problems.

3D Surrogate Modeling for Design Optimization

Data-driven surrogates built on geometry-aware neural networks (point clouds, meshes, graphs) that replace costly simulations and enable scalable design optimization.

Proposed Frameworks

Agentic AI
LLM-Guided Adaptive Penalty

An LLM adaptively tunes penalty parameters inside Analytical Target Cascading for constrained design.

Agentic AI
AutoATC

Automatically decomposes complex multidisciplinary design problems using LLM-based agents.

Agentic AI
DialogCAD

A dialogue-driven CAD generation framework powered by a multi-agent system over the Model Context Protocol.

3D Surrogate Modeling
Point-DeepONet

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

3D Surrogate Modeling
Physics-Constrained GNN

Spatio-temporal prediction of drop impact on OLED display panels with physics constraints.

3D Surrogate Modeling
BMO-GNN

Bayesian mesh optimization that boosts engineering performance prediction with graph neural networks.

See Publications See Projects