{
  "schema_version": "1.1",
  "id": "auto-e7ee7f4465",
  "slug": "lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465",
  "url": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465",
  "title": "Lessons from Studying Every Memory System — Shlok Khemani, Independent",
  "why_included": "Make agent memory inspectable and editable, with conflict detection and an explicit update cadence to prevent false beliefs from persisting.",
  "summary": "Persistent memory is a compute and product tradeoff, not just retrieval. Profiles need conflict detection, visibility, editing, and deliberate update cadence before agents can rely on them.",
  "practical_implication": "For agent memory, choose update frequency and profile size as an explicit compute budget. Make stored beliefs inspectable and editable, preserve source context where possible, and detect uncertainty or contradictions before a profile silently steers future work.",
  "agent_context": "ChatGPT and Claude have converged on running user profiles plus tools for retrieving past conversations. The observed implementations differ: ChatGPT's profile is about **4,000 tokens** and updates every few days, while Claude's is about **1,000 tokens** and updates every 24 hours.\n\nFor agent memory, choose update frequency and profile size as an explicit compute budget. Make stored beliefs inspectable and editable, preserve source context where possible, and detect uncertainty or contradictions before a profile silently steers future work.\n\nThe Turkey example shows the core failure: conversations about possible travel were condensed into a trip that never happened. Neither profile summaries nor retrieval alone solve missing evidence, cross-product context silos, or reconciliation with email, calendars, and other sources.",
  "source": {
    "name": "AI Engineer",
    "url": "https://www.youtube.com/watch?v=5ZGyKWjQDr0",
    "published_at": "2026-08-12T00:00:00.000Z"
  },
  "source_class": "video",
  "content_type": "Video",
  "layer": "agent",
  "domains": [],
  "topics": [
    "agent-memory",
    "context-engineering",
    "retrieval"
  ],
  "verification": {
    "status": "source_linked",
    "label": "Source Linked",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [],
  "connected_context": null,
  "lifecycle": "New",
  "published_at": "2026-08-12T00:00:00.000Z",
  "modified_at": "2026-08-12T00:00:00.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465",
    "json": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465.json",
    "markdown": "https://feed7.dev/p/lessons-from-studying-every-memory-system-shlok-khemani--e7ee7f4465.md"
  }
}