You understand this is more than that right? This is saying, next time you need to analyze a big dataset, tell your AI to load it in DuckDB, and if it needs to run a query that SQL cannot do, tell it to do what they did in this blog, and it'll have the query run using Clojure super fast.

It's not really about querying a DB from Clojure, well not in the boring sense of how do I run some SQL over data in an existing DB.

What this is supposed to replace is say a big Spark Cluster for processing/querying large datasets.

I’m not so sure - architecture matters, efficiency matters, extensibility matters, and all are impacted by language choice.

If we had developed our tech in Python we’d be a world of hurt now as we scale.

These takes are not interesting, relevant, or helpful.

Then perhaps this isn't for you.