The Problem
The internet's Aesthetics Wiki catalogs hundreds of subcultures but is unstructured, inconsistent, and unqueryable — unusable for researchers and designers who need clean, prompt-ready data.
The Solution
A clean, queryable HuggingFace dataset of ~400 aesthetic subcultures scraped and normalized with Python and Pandas, enriched with tags, color palettes, and iconography. 239 downloads as of Nov 2025 by designers, researchers, and AI practitioners.
Tools
Web ScrapingData CleaningPython + PandasNotionHugging Face Datasets
Architecture / Flow