Hello, I'm Michael.
I build reliable, high-quality products with great UX.
This is my home base: projects I'm proud of, writing that captures what I learn, and a simple way to connect.
Feature Projects
A few recent highlights. For more details, please navigate to the Project page.

App Review Intelligence Platform
Full-Stack DeveloperA full-stack platform that collects, tags, and analyzes app store reviews with AI. It can benefit product and development teams to spot pain points, track sentiment over time, and query user feedback through a RAG-powered chatbot.
Handled full project lifecycle: system design, API architecture (FastAPI), state management, error handling, and deployment.

Impact Data and Evidence Aggregation Library (IDEAL)
DeveloperA project that builds an open-access platform for researchers and policymakers to search, compare, and visualize policy intervention evidence across countries. It facilitates meta-analyses, systematic reviews, and evidence standardization.
Contributed to web and app design, full-stack development from 0 to 1, data visualization, and mini toolkit components including a comparative review Shiny app and ML-based screening tools.
Writing Samples
Each piece starts with a question: how do words move markets, who really owns the data, or what makes a system reliable? Selected work across languages and topics.
NLP, LLM, AI & Automation
How can we automate workflows with AI? · What tools make data pipelines reliable?

AI-Driven Data Viz for Psychology of Poverty Literature Review
LLM classification and visualization for two key labels about context and mechanisms

Automating GitHub Projects V2 with Python
GraphQL API patterns, GitHub Actions, and idempotent automation

Building Interactive Dashboards with Streamlit and Plotly
Real-time project tracking with Google Sheets, Pandas, and Plotly
Computational Social Science
How do words move markets? · What shapes human behavior at scale?

Decoding Children's Adulthood Income: Parents, Races, and Neighborhood
Intergenerational mobility, race, and neighborhood effects on income

Evolution and Guidance Power of FOMC Meeting Documents
Text analysis of Fed communications and their link to Treasury yields

Misinformation and Deplatforming on Twitter: A Replication
DID and regression discontinuity on post-Jan 6th deplatforming effects

Urban Mobility and Distribution of Crimes in NYC: Beyond Neighborhood
Network analysis of how community mobility shapes crime diffusion across New York City
Data & Public Policy
Who owns the data? · How can data help decision making?

Regulating the Use of Biometrics in the US and the UK: A Comparative Study
Comparative analysis of biometric data regulation across two jurisdictions

Mid-to-High Frequency Economic Indicators and GDP Growth: A Shenzhen Case Study (In Chinese)
Econ indicators and GDP growth prediction

Public Data Authorization and Operation: Practices and Frameworks (In Chinese)
Public data licensing models across China