Yes, I heavily rely on such a device, as most of my devices are asynchronously online, e.g I have a laptop for work and a home PC. When I’m at home I do my work on the PC, then turn it off. The next day I usually turn on the laptop when I arrive at work. The laptop then starts syncing the changes from the previous day from the central server.
A bunch of computers - android, windows, linux in all sizes (from smartphone/tablet over server racks to tower pc) that is connected in a wild mix of star and mesh topology using syncthing. Some folders are shared between all, some are only shared only between some, depending on needs and storage.
I also have a second set of instances running (on some machines) that share folders to my friends. I don’t want my friends to interfere with my “home” syncthing setup, so I’m using segregated syncthing instances for that matter, to form an entirely separate cluster that is totally unaware of the other.
Not really sure if it defines as cloud, but yes I store encrypted backups on a remote storage-only server using BorgBackup.
I already mentioned BorgBackup. I do additional local backups via cold/offline storage disks. I send data to these disks using FreeFileSync (I connect the backup disks in regular intervalls, copy stuff over, then disconnect and store the backup disks again). For actively syncing data, my day-to-day tool really is Syncthing.
I also have a handful of config files that need to be synced between two servers, but change very rarely. Those configs also need to by owned by a system user, so using syncthing for this feels overkill. Instead I use a combination of autossh (automatic ssh tunnel) and Lsyncd (inotify-trigger-based rsync) to sync these few kilobytes of very-low-frequency files.