This is exactly the direction I’m thinking about.
Right now SysAI is mostly workflow/context aware within the current session, but not yet maintaining a persistent operational baseline across sessions.
But I think you’re completely right that “drift from known-good state” is where generic AI chat completely falls apart for real infrastructure work.
The more I build this the more I realize:
chat is useful for discovery, but persistent operational context is probably the real product layer.
That’s actually why I recently added an issue for an operational context memory system instead of going down the “autonomous agent” route.
