Vibe Design: Human-in-the-loop AI Agents for UI Design with Large Language Models

New tectonic shifts are looming in the UI design process due to advances in LLMs. These shifts have led researchers to adopt LLM-based approaches to develop design systems. However, such systems often focus on narrow functions and rely on code-based data that limits real-world applicability. We propose Vibe Design, an agentic framework that enables designers to co-create and evaluate prototypes with the LLM-based agents. The framework consists of a design agent that supports UI creation and a user testing agent that generates usability feedback through simulated user interviews, guided by the Persona-Scenario-Goal methodology. We also incorporated a human-in-the-loop mechanism to enhance the reliability and quality of the LLM-generated responses. We successfully generated functional prototypes and extracted user requirements from diverse personas. This study provides insights into the use of LLMs for both UI generation and user feedback collection, offering practical implications for future design workflows.

Authors

Publication Details

Journal
ScholarSpace (University of Hawaii at Manoa)
Published
2025-12-23
Primary Topic
Persona Design and Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Vibe Design: Human-in-the-loop AI Agents for UI Design with Large Language Models

Ji Sun Hwang, Juyoung Kang
ScholarSpace (University of Hawaii at Manoa)
Persona Design and Applications
article

Vibe Design: Human-in-the-loop AI Agents for UI Design with Large Language Models

Ji Sun Hwang, Juyoung Kang
article en

Abstract

New tectonic shifts are looming in the UI design process due to advances in LLMs. These shifts have led researchers to adopt LLM-based approaches to develop design systems. However, such systems often focus on narrow functions and rely on code-based data that limits real-world applicability. We propose Vibe Design, an agentic framework that enables designers to co-create and evaluate prototypes with the LLM-based agents. The framework consists of a design agent that supports UI creation and a user testing agent that generates usability feedback through simulated user interviews, guided by the Persona-Scenario-Goal methodology. We also incorporated a human-in-the-loop mechanism to enhance the reliability and quality of the LLM-generated responses. We successfully generated functional prototypes and extracted user requirements from diverse personas. This study provides insights into the use of LLMs for both UI generation and user feedback collection, offering practical implications for future design workflows.

ScholarSpace (University of Hawaii at Manoa)
Openalex Percentile: Top 34%
Persona Design and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.