Value context and insight. lkm@koreatimes.co.kr
KAIST workshop to showcase AI agents for HBM design automation
Kim Joung-ho, third from left, an electrical engineering professor at the Korea Advanced Institute of Science and Technology (KAIST), lectures students at the university’s campus in Daejeon. Courtesy of KAIST
By Lee Kyung-min
The Korea Advanced Institute of Science and Technology (KAIST) will showcase on Friday artificial intelligence (AI) agents capable of automating high-bandwidth memory (HBM) design, offering a glimpse into the future of semiconductor research.
KAIST's HBM Design and Research Automation Workshop Using the OpenClaw AI Agent will run from 8 a.m. to noon and highlight practical applications of the laboratory's in-house OpenClaw AI Agent platform.
Graduate student researchers from the laboratory, led by KAIST electrical engineering professor Kim Joung-ho, will demonstrate practical applications for these AI agents. Their presentations will focus on automating chip design and simulation, optimizing document management and streamlining semiconductor research workflows.
The first session, from 8 a.m. to 10 a.m., will introduce the OpenClaw AI Agent platform and its use with the Model Context Protocol (MCP).
Presentations will cover AI agents for package design and analysis using KiCad MCP, semiconductor power delivery network simulation, eye diagram dataset construction and equalizer optimization, and Ansys electromagnetic simulation automation using PyAEDT.
Researchers will also demonstrate knowledge base construction using Docling and LLM Wiki, collaborative AI agent chatrooms and automated patch antenna design in the Ansys HFSS environment.
The second session, from 10:15 a.m. to 11:45 a.m., will focus on AI agents that support researchers' daily work, including portfolio management, automated stock trading through securities firm APIs, paper keyword collection, Google Calendar-based presentation draft generation and Linux server management.
The laboratory said videos of the presentations will be released later through its website.
This article was published with the assistance of generative AI and edited by The Korea Times.