lmk any thoughts or comments!

Hello!

A few months ago I put out a small pgvector vector visualizer: https://news.ycombinator.com/item?id=42457782 and it got more interest than I expected.

dbSurface is intended to be a cleaner, production ready tool. It allows you to create 2d projections of postgres tables with vectors, query across the projections, and check query precision. Hopefully this allows for quick iteration in query/embedding strategies.

It uses a hybrid hosted setup: you run dbsurface's docker image locally and query your database directly from your machine, but it offloads dimensionality reduction to a GPU powered cloud worker so you can create a projection of a 1 million row table in ~10mins.

Would love to hear any feedback, ideas, or questions!

It lets you visualize any column with type "EMBEDDING", and I think the only way to get that is through pgvector/pgvectorscale.

lmk if anyone has any thoughts...if I could go back I may have not gone through Electron

Doing dimensionality reduction locally posed a few challenges in terms of application size--the idea was that by analyzing just a few thousand randomly sampled points you can get an idea of your data through a local GUI where you interact with your data and see some correlated metadata.

Not sure if there's too much need for an individual GUI to go along with Postgres as a VectorDB, maybe people just do analysis separate from a normal "GUI"? But maybe not.

What you think?

Just some fast feedback, I can't copy & paste in the connection url input form. On a mac.

Once loaded, I get the error "Table must contain a UUID column for vector visualization."

I'm assuming it's trying to find an ID column for grouping? Can we manually specify this? My ID columns are varchars.

Same here. I'm using langchain which creates a varchar id column. It also has different collections on the same table.