I only have one: (Py)Spark. I do a lot of ETL and it has solved most of Pandas arbitrary data type limits (ex. dates). Rust based alternatives are still not stable enough in my experience.