> Timescale lately released the pgvector extension, that turns your PostgreSQL into a vector database.
I don't think this is accurate and smells like an LLM hallucination to me.
From the Timescale/Tiger Data _pgvectorscale_ project's README:
> pgvectorscale builds on pgvector with higher performance embedding search and cost-efficient storage for AI applications.
I think this is where the confusion originates. I believe pgvector is primarily Andrew Kane (@ankane) and a cadre of OSS contributors.
As an aside, I've used Timescale/Tiger Data products and was very happy with them and their support. Their team was very engaged and responsive to all of our questions. They also fixed a pretty gnarly indexing bug I uncovered in pgvectorscale in an impressively short amount of time.
Thanks for the kind words.
And yes, Andrew Kane (et al) are the people to thank for pgvector.
We (Tiger Data) developed pgvectorscale and pg_textsearch (and timescaledb, and some others)