Could this be solved by (periodic) snapshotting into a system with good support for indexing in nested documents, for an initial eligibility search, then perhaps re-snapshotting with live updates, mixing in recently updated documents by the current user for read-your-writes optics, as a final in-memory filter?
Yeah but the syncing might be more trouble than it's worth, I was thinking of doing this. I know there are sync engines like ElectricSQL, maybe you could merge that with a CRDT but not sure.
