DSS does not have support for Presto at the moment.
You can try installing the JDBC driver for Presto as explained here: https://doc.dataiku.com/dss/latest/installation/jdbc.html and then use the "Other databases (JDBC)" dataset, but this generally only provides limited read-only support, and may not work.
Prestogres (https://github.com/treasure-data/prestogres/) might be better choice at this stage. It is based on PLPython and it does replicate the Presto dataset into PostgreSQL tables and uses UDFs to select directly from Presto. The setup looks straightforward as well.
I will try it out and see how it goes. We might be able to remap some of the column definitions in PostgreSQL.
It would be great if Presto can be on the Roadmap. It is a "distributed SQL query engine designed to query large data sets distributed over one or more heterogeneous data sources".