Geospatial AI is a broad field that encompasses the application of AI to geospatial data to solve real-world problems. Our research spans multiple domains including localization, remote sensing, ecology, webcam networks, social media analysis, transportation, agriculture, and LiDAR processing. Recent work includes Earth Embeddings for AI-centric representations of our planet, retrieval-augmented neural fields for multi-resolution geo-embeddings, active geospatial search frameworks, and unified embedding spaces for ecological applications. We develop novel algorithms that integrate diverse geospatial data sources—from satellite imagery to ground-level photos—to address critical challenges in environmental monitoring, urban planning, transportation safety, and conservation.