I gave this a try. Looks neat however one thing I noticed is that it does not seem very performant in querying parquet files. Take this simple query for example: 'select * from ./order.parquet limit 5;' It takes roughly 38seconds to execute on my machine (I ran it a few times). The order file is about 300mb with 7.5m records in it. To give some comparison DuckDB can run this query in less than a second (while also directly accessing the parquet file).