Postgres is not typically considered to "scale well," but oftentimes this is a statement about its tablespaces more than anything; it has foreign data[4] API, which is how you extend Postgres as single point-of-consumption, foregoing some transactional guarantees in the process. This is how pg_analytics[5] brings DuckDB to Postgres, or how Steampipe[6] similarly exposes many Cloud and SaaS applications. Depending on where you stand on this, the so-called alternative SQL engines may seem like moving in the wrong direction. Shrug.

Maintainer of pg_analytics (now part of pg_search) here. I 100% agree that the statements against Postgres are often exaggerated. In practice, we see both the smallest and the largest companies "just use Postgres" while mid-scale companies often overthink their solution.

That said, there are indeed phenomenal "alternate" SQL engines. I've seen many users see great success on tools like ClickHouse, which ParadeDB is not yet competitive with, and sometimes (dare I say) even Elasticsearch. As for whether this one is one of them... That I couldn't say