Durable Objects for agent state is the correct move for low-latency memory. I've been building these exact patterns as a TypeScript first class memory primitive for AI agents: https://github.com/ereztdev/langmem-ts it handles the extraction/recall loop without the Python-heavy overhead of LangMem or Mem0. Pure TS, no wrappers

The profile as a memory primitive is pretty interesting, how do you scale it for a scoped access/hierarchical setting? (for example, overall memory of processes => customer specific => project specific memory)