UT Austin Geoscience Hackathon: `AlphaEarthHack` team

VectorScope: Similarity Search with Google’s Satellite Embedding V1 empowers users to map surface types by drawing sample points or loading point assets. The app computes cosine-similarity heatmaps of 64-dimensional embeddings for any year, offers threshold tuning and heatmap toggling among other functionalities

Binary classification experiments to interpret Google AlphaEarth Foundation embeddings across ESA WorldCover land cover classes. Part of the study "What on Earth is AlphaEarth?" — 130,000+ experiments using Random Forest, XGBoost, LightGBM and progressive ablation across 64 embedding dimensions.

Unsupervised agricultural anomaly mapping using AlphaEarth satellite embeddings, K-Means baseline clustering, and CHIRPS rainfall validation.

Near-term wildfire risk forecasting platform for California using H3 geospatial indexing, NASA FIRMS, AlphaEarth embeddings, weather features, baseline ML models, and a map-ready dashboard.