Edge caching strategies for personalized feeds

Personalization and caching are supposed to be enemies: the moment a response depends on who is asking, the cache hit rate falls off a cliff. But most feeds are less personal than they look.

By splitting the response into a shared, cacheable skeleton and a thin per-user overlay — and choosing cache keys that reflect that split — the edge can serve the expensive part once and assemble the cheap part per request.