Overview
In addition to the API, PropertyLayer offers parcel data as flat file exports: full county or statewide datasets delivered as structured files. Flat files are the right choice when you need to process large volumes of parcels, run offline analysis, or build embeddings and AI pipelines.API vs. flat files
Use the API when you need real-time access, geographic queries against user-submitted polygons, or per-parcel lookups in a live application.
Use flat files when you need more than 5,000 properties/month, are benchmarking data quality before committing, building embeddings for an AI workflow, or loading parcel data into your own data warehouse.
AI and ML workflows
Most AI and ML work wants the whole corpus offline rather than one property at a time, which is what flat files are for. Typical uses are batch embedding generation over full parcel records in Databricks or Spark, vector indexes built over the text fields (owner, address, legal description), and semantic search that embeds a query and retrieves similar parcels. One caveat: flat files are not the same shape the API returns. A response carries sections such asownership and valuation_current that the flat files do not. Null behavior is the same in both: a missing value means the county did not publish it.