The project is still fairly new and not close to production tbh. I'd actually recommend the simplest option: just write them to Parquet on S3 and query with pl

The project is still fairly new and not close to production tbh. I'd actually recommend the simplest option: just write them to Parquet on S3 and query with plain DuckDB. Or you could use Ducklake - https://ducklake.select/ https://ducklake.select/