Explore large-scale geospatial indexing and analysis using Apache Spark in this 23-minute conference talk by Databricks. Delve into the challenges of processing geospatial data at scale, examining open-source frameworks like Apache Sedona and its improvements over conventional technology. Learn about spatial data structures, formats, and indexing techniques such as H3. Discover how these components integrate into a cloud-first architecture utilizing Databricks, Delta, MLFlow, and AWS. Examine practical examples of geospatial analysis with complex geometries and spatial queries. Gain insights into augmenting analysis with machine learning modeling, human-in-the-loop annotation, and quality validation. The talk covers topics including spatial indexing, use cases, SQL queries, spatial joins, geometry overlap, and overall architecture, providing a comprehensive overview of large-scale geospatial data processing and analysis techniques.