Auto Create and Sync Vector Embeddings in 1 Line of SQL (pgai Vectorizer)

[@PeeyushU]

This is great! For blobs managed by a lifecycle policy on s3, I guess, a CF can insert/archive a meta record in the content table, and the vectorizer update/archive embedding periodically

[@myctoai]

Nice. What is the postgres management desktop app you are using there? Thanks!

[@McAko]

PopSQL, acquired by Timescale

[@myctoai]

@@McAko Cool nice I'm going to try that out, appreciate you.

[@awakenwithoutcoffee]

definitely going to test with this. Postgres < AI

[@Omar334-asd]

Hi! Is it possible to use embeddings to summarize a long description into three sentences?

[@louisbertson]

That is amazing. Thank you

[@TimescaleDB]

Glad you enjoyed it!

[@barefacedquestions]

Thank you for the great presentation. So with pgai, we can bypass the step of using pgvectorizer to embed PostgreSql data?

[@TimescaleDB]

pgai Vectorizer makes it easy to embed data. So you don't have to run your own embedding creation and sync pipelines.

[@dellanmuchengapadare3792]

I can imagine how easy it can be to build a resume screening workflow 😅

[@serychristianrenaud]

thanks

[@TimescaleDB]

You're welcome!

[@SneheshThalapaneni]

@Timescale what's the name of the db client at kzbin.info/www/bejne/kKCmY4uPpNFpkM0

[@alissonmorais5463]

Not sure, but maybe Azure Data Studio?

[@RahilSondhi]

It's PopSQL!

[@SneheshThalapaneni]

@@RahilSondhi Thanks