University of California, Irvine | 2024 – Present
• Conducted a systematic literature review on women's health in HCI, mapping research trends and identifying design gaps.
• Designed and deployed a mixed-method survey with 249 U.S. women to study inter-generational sharing of genetic health experiences.
• Analyzed survey data to surface narrative patterns and barriers, informing speculative design opportunities for genetic health sharing, such as AI-driven storytelling.
Northeastern University | 2021-2023
• Conducted interviews with 19 older adults and healthcare providers to uncover communication barriers and design opportunities for LLM-powered systems.
• Designed and developed Talk2Care, integrating a voice assistant (Alexa + Python Lambda) for patients and an information dashboard (Figma, React) for providers with GPT-4.
• Ran mixed-method user studies (Qualtrics SUS survey + interviews), demonstrating strong usability and surfacing design implications for LLM-mediated communication.
• Tested LLaMA-2 (7B/13B/70B) models on Amazon BedRock and SageMaker, exploring customization, personalization, and scalability for healthcare contexts.
• Facilitated participatory design sessions with cancer care providers and patients; developed a GPT-4–enabled monitoring system, RECOVER, and conducted multi-stage deployment studies and mixed-method evaluation.
Zhejiang University | 2024
• Programmed Python scripts for the NAO social robot and designed a between-subject experiment on generative AI–driven embodied learning.
• Applied quantitative analysis (learning and emotion scales, ANCOVA) to evaluate knowledge acquisition and emotional outcomes.
• Led qualitative analysis for survey data using iterative coding and thematic analysis, identifying 23 axial codes and 6 themes (inter-rater reliability k=0.85).
University of Michigan, Ann Arbor | 2022–2024
• Prototyped empathetic exercise coach chatbots (scripts + Figma) to study long-term user–chatbot relationships; conducted pilot studies (n=6) with surveys and interviews; analyzed data from Amazon Mechanical Turk to reveal user preferences.
• Developed AI-driven Just-In-Time Adaptive Interventions (JITAI) using technology probes to promote physical activity.
University of Michigan, Ann Arbor | 2022–2023
• Designed 3D graphics for AR/VR environments on Meta Quest for the shape and flow of languages, prototyping and testing experiences using Unity and Unreal Engine.
• Developed and deployed the final immersive environment via Mozilla Hubs.