I still feel like I’m living in the future using Paperless though I feel I must use it differently to everyone else as i haven’t felt the need to automate everything as it doesn’t always tag things to match my own specific tagging system.
For those that do automate everything, how do you overcome naming of documents? I want the scanned documents to have human readable names rather than the naming format of my scanner (or phone on occasion) or the naming format of XYZ company emailing me a document because in a catastrophe situation I want to be able to easily find & retreive vital/important .pdf docs from a backup rather than having to think about replacing equipment to set up Paperless again (in my instance aside from nightly Kopia snapshot backups I also export Paperless backups on device plus a secondary backup is automatically moved to a separate drive on my network & a third copy is encrypted & stored off site).
I run Nextcloud + Memories plugin on a Pi5 8GB with NVMe with a fanless Argon Neo case. Fast & very stable. One user but it gets quite heavy use.
An alternative you might prefer is a Beelink mini PC which I’m runnin with SSD’s. I went for a Beelink EQ14 which i use for stuff like Paperless & Immich plus several other self host softwares. Picked it up cheap about a year ago before the AI nonsense pushed prices high. Frugal with electricity & more powerful than Pi. Ships with Windows which I ditched for Ubuntu Server. I’ve found it to run like a dream.