The author of the text is building a COVID-19 geosearch site to explore case counts using a map interface. They select a data source from John Hopkins University's aggregated COVID-19 dataset and choose to consume it through different interfaces, including flat files, MongoDB, REST API, and GraphQL API. The goal is to create an index in Algolia that provides fast and accurate search results. After selecting the data source, they shape the data for their index by trimming records down to the simplest form that still provides the right balance between performant search and useful results. They send their normalized data to the Algolia Search API using the Python SDK. The author discusses various strategies for keeping the index up-to-date as the underlying data changes, including clearing and rebuilding the index at fixed intervals or performing incremental updates to individual records or fields of data. Finally, they focus on building their front end using the Algolia React InstantSearch library and Google Maps.