Shu Yang

Designer & Technologist merging urban science with creative innovation. Expertise in UI/UX, branding, data visualization, and machine learning. Projects experience in AI, AR, and decentralized gaming. Passionate about user-centric solutions bridging design and tech.

SomniBot: AI-Powered Personalized Sleep Assistant

SomniBot is an AI-driven sleep assistant that provides personalized, science-backed sleep guidance. Using FHIR for secure health data handling and Retrieval-Augmented Generation (RAG) for accurate responses, it offers customized sleep insights based on user input and electronic health records (EHR). With features like a self-assessment quiz, intelligent sleep analysis, and expert-backed recommendations, SomniBot helps users improve their sleep quality while ensuring privacy and reliability. Future enhancements include a dark mode for better nighttime usability, an FAQ section for common sleep concerns, and expanded EHR capabilities.

Houston's Healthcare Divide: Mapping Access and Inequality

This study delves into the spatial accessibility of healthcare services in Houston by employing Geographic Information System (GIS) network analysis. Through the examination of actual road networks and travel times, it reveals the uneven distribution of healthcare resources in Houston and how this disparity impacts residents' access to healthcare.

Economic Toll of Flooding: Florida Case Study

This project predicts flood risk and economic losses in Florida using GIS and machine learning. By integrating physical, demographic, and socioeconomic data, we model flood vulnerability and economic impact at the census tract level.

CLT Legal AI Assistant: Intelligent Chatbot & Multilingual Platform

An AI-powered legal support tool for Community Land Trusts (CLTs), leveraging machine learning to parse legal datasets and scholarly articles. Iteratively enhanced across three versions, featuring:
V1.0: Basic chatbot responses, interactive map, and localized law integration.
V2.0: Personalized query tracking, user accounts, and expanded state/municipal law databases.
V3.0: Multilingual support, global legal datasets, and context-aware insights for actionable solutions.
Delivered intelligent content extraction and dynamic user engagement tools to streamline CLT policy drafting and compliance.

Facial Pixel Mosaic: Real-time Generative Art with Eye Tracking

An interactive digital artwork combining face detection (via ml5.js FaceMesh) and pixel manipulation. Captures live video to create dynamic mosaic portraits using color-mapped grayscale thresholds. Features eye-triggered regeneration - resets color patterns when eyes close (landmark height 7px). Click interaction allows manual refresh. Demonstrates creative coding with computer vision and real-time graphics synthesis.

What I do

Design Expertise

  • UI/UX design, branding, visual identity
  • Figma, Adobe Suite, Cinema 4D

Technical Proficiency

  • ML development, geospatial analysis, and data visualization
  • Languages: Python, JavaScript
  • Tools: ArcGIS, OpenAI API

User Research & Analytics

  • 50+ usability tests conducted
  • 1,000+ market data points analyzed

Get in touch