GeoEstateChat is a research project exploring how LLMs use natural language to connect real estate decision-making with multi-scale geospatial analysis. It investigates how user intent can mediate between spatial data, analytical workflows, and real-estate reasoning.

Through Exponential Regression and Logistic Regression, the project predicts how much we can find out about a city, only through the analysis of Street Block Geometry:

The project uses K-Mean clustering for grouping Manhattan's residential buildings with different themes based on 11 different residential dataset.

A Python package that helps row-wise merging of messy tables reliably. It prepares tables for merging, finds matching columns, suggests merge keys, and explains why merges succeed or fail.

Using Rhino Grasshopper and C#, the project calculates and maps the walking time from each building in Manhattan to its nearest subway station, based on shortest-path analysis of the pedestrian street network.

Architectural design project that redefines Hong Kong's cruciform tower typology through in-depth parametric analysis of building codes, design grammar, and view corridors.

Architectural design project reimagining Hong Kong's podium tower through generative iterations, exploring spiral geometries and walkability analysis.