Exactly. We use Supabase too but are at a scale where it just made sense to use a second, dedicated vector db (Pinecone) than to bloat our Postgres db that has a completely different workload
I'd really love to know what kind of insane scale justifies that tradeoff...
Of course if I sounded incredulous it's because I didn't think you had that scale, and it sounds like I was correct?
We don't also "pipe" our data to Supabase, we use a couple different data stores depending on the best use case. For example we also use R2 and Durable Objects.
Just because you have a hammer doesn't mean everything is a nail.
Also don't you think it's funny calling out somebody else's tech choices when you have zero insight into it and when it's worked out perfectly for us?
By the way, how many TBs of vector data are you storing in Postgres and needing to retrieve with minimal latency?
Literally irrelevant lol
For basic CRUD we use the Supabase endpoints directly but none of that involves querying a vector db :P