AI Product Manager · LLM / RAG / Prompt Engineering · HCI · Conversational AI · UX Research
Hi, I'm Zoe Zhu, an aspiring AI Product Manager with a background in human-computer interaction, instructional design, and AI system evaluation. I combine technology, research, and product thinking to build intelligent experiences that are human-centered and grounded in real user needs.
I am currently pursuing my M.A. at Teachers College, Columbia University, with a GPA of 4.0/4.0 and a focus on AI learning systems, human-computer interaction, and organizational learning.
My experience spans AI product evaluation for the LiLo NLP chatbot at UCI daplab, AI tool assessment for learning and development at the United Nations, and independent AI product development, including Haoyun AI, a Chrome extension built with Codex, Figma-to-code workflows, and LLM APIs.
My product experience includes AI evaluation frameworks, user research synthesis, data-driven iteration, and hands-on LLM application development.
AI Product Management · AI Product Evaluation
LLM / RAG / Prompt Engineering · Conversational AI
Human-Computer Interaction · Human-AI Interaction
UX Research · User-Centered Design
Instructional Design · AI Learning Systems
UCI daplab, Dr. June Ahn Lab
Participated in AI product evaluation and UX research for LiLo, a conversational chatbot built with NLP and GPT-2 to help high school graduates navigate the college enrollment process. The project partnered with NYU and CUNY and supported approximately 20,000 students.
My work included user research data collection, Excel-based analysis, and Figma UI/UX optimization.
Live Demo: https://haoyun.figma.site
Built independently using Codex, Figma-to-code workflows, and LLM APIs. The extension uses LLM-based field matching to auto-fill resume forms across LinkedIn, Zhaopin, and BOSS Zhipin.
Stack: JavaScript · HTML · CSS · LLM API · Prompt Engineering · Chrome Extension Development
Live Demo: https://roomate.figma.site
Built a 6-dimension behavioral compatibility assessment tool with 0-100 scoring and persona mapping. Users answer 20 lifestyle questions and receive structured, explainable compatibility feedback.
Prototype: https://emotion-support.figma.site
A Next.js web app exploring privacy-first AI emotional support through conversational interaction, emotion recognition, and mood trend visualization.
AI & Product:
AI Product Evaluation · Prompt Engineering · LLM / RAG / Token · Conversational AI · User Research · Needs Analysis · Competitive Analysis · Data-Driven Decision Making · Product Thinking
Design & Research:
Figma · UX Design · Interaction Design · Conversational UX · Prototyping · Instructional Design
Technical:
Python · JavaScript · HTML/CSS · Chrome Extension Development · Vibe Coding with Codex · LLM API · Excel Pivot Tables · VLOOKUP · Google Workspace · Gradio · LMS Platforms
Email: zz3378@tc.columbia.edu
Phone: +1 3477328050 / +86 13336816750
GitHub: https://github.com/zoezhuy
This GitHub profile documents my projects and explorations at the intersection of AI product thinking, human-computer interaction, LLM applications, and user-centered design.