The heiURBAN corpus is compiled by collecting and organizing language data from various modalities such as text, speech, image, and video within one digital interface to document the dynamic and rapid gentrification of Brooklyn, capturing how urban spaces are transformed into lived, meaningful places. By integrating visual, textual, and digital traces, we aim to trace patterns of urban change and linguistic diversity simultaneously. This approach addresses the methodological challenges of collecting and interpreting multimodal data, while highlighting the reflexive role of the researcher in digital sociolinguistics and urban ethnography.