Lots of companies started to create AI Assistant based on LLM and vector database.

Many questions appeared with those new technologies:

* Are existing Helm charts able to support or include vector plugins? Do we have to rebuild and maintain those images?
* Are we able to backup and monitor those databases?
* Are pure vector database the real target in front of historical databases (mongodb, postgres, etc.) using RAG method?

We would like to share our experiences about different implementations of databases on Kubernetes with our different customers.