{
  "schema_version": "1.1",
  "id": "s8:https://www.youtube.com/watch?v=O84lhGc1OOI",
  "slug": "if-we-want-them-to-do-knowledge-work-design-them-as-knowledge-agents-ben-1f64i2y",
  "url": "https://feed7.dev/p/if-we-want-them-to-do-knowledge-work-design-them-as-knowledge-agents-ben-1f64i2y",
  "title": "If we want them to do Knowledge Work, design them as Knowledge Agents — Benjamin Clavié, Mixedbread",
  "why_included": "Knowledge agents need intent decomposition, specialist researchers, and multiple retrieval primitives; copying a coding-agent loop fails when evidence is ambiguous and context-dependent.",
  "summary": "Clavié argues that non-code knowledge lacks durable identifiers and explicit definitions: a phrase such as “30 days” can represent several unrelated obligations. Search therefore begins with intent and conditional context, not a grep-ready target.",
  "practical_implication": "Design the main agent like a professional lead: decompose the question, send bounded research to **specialist subagents**, and combine their memos. Give the harness multiple primitives—**BM25, semantic search, and multimodal PDF search**—and teach it when each applies.",
  "agent_context": "Clavié argues that non-code knowledge lacks durable identifiers and explicit definitions: a phrase such as “30 days” can represent several unrelated obligations. Search therefore begins with intent and conditional context, not a grep-ready target.\n\nDesign the main agent like a professional lead: decompose the question, send bounded research to **specialist subagents**, and combine their memos. Give the harness multiple primitives—**BM25, semantic search, and multimodal PDF search**—and teach it when each applies.\n\nThe cited benchmark result says improved search orchestration reduced the human-agent mistake gap by **40%**, but the talk omits enough setup detail to generalize that number. Better orchestration also cannot recover evidence absent from the corpus or retrieval tools.",
  "source": {
    "name": "AI Engineer",
    "url": "https://www.youtube.com/watch?v=O84lhGc1OOI",
    "published_at": "2026-09-16T15:00:39.000Z"
  },
  "source_class": "video",
  "content_type": "Video",
  "layer": "agent",
  "domains": [
    "research"
  ],
  "topics": [
    "harness-engineering",
    "subagents",
    "retrieval"
  ],
  "verification": {
    "status": "source_linked",
    "label": "Source Linked",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "The cited benchmark result says improved search orchestration reduced the human-agent mistake gap by **40%**, but the talk omits enough setup detail to generalize that number. Better orchestration also cannot recover evidence absent from the corpus or retrieval tools."
  ],
  "connected_context": {
    "meaning": "This sharpens knowledge-agent design around ambiguity: the lead agent must infer intent, decompose research, and choose among retrieval modes rather than merely distribute queries. It supports bounded specialist subagents while conflicting with fixed multi-agent chains that lose shared context. The reported mistake-gap reduction is evidence for orchestration, but remains too underspecified to generalize.",
    "corpus_size": 807,
    "generated_at": "2026-09-18T10:06:25.197Z",
    "connections": [
      {
        "title": "Where RL Will Take Search — Maximilian-David Rumpf, SID.ai",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=iJVxxxHM_Oc",
        "feed7_url": "https://feed7.dev/p/where-rl-will-take-search-maximilian-david-rumpf-sid-ai-1jeb1sw",
        "reason": "Both isolate retrieval in bounded specialists to protect the lead agent’s context; this Signal adds that those specialists need multiple search primitives selected according to intent and evidence type."
      },
      {
        "title": "Why We Killed Our Multi-Agent Pipeline — Subbiah Sethuraman and Abhilash Asokan, ZS Associates",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=u6jJcIFDLE4",
        "feed7_url": "https://feed7.dev/p/why-we-killed-our-multi-agent-pipeline-subbiah-sethuraman-and-abhilash-a-0fmz3z3",
        "reason": "The failed fixed pipeline supplies the architectural boundary for this proposal: one lead retains end-to-end ownership while subagents handle bounded investigations rather than passing the whole task through a chain."
      },
      {
        "title": "It’s Tokens All The Way Down: How RLMs are Different — Kevin Madura, AlixPartners",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=xo68uCibfm8",
        "feed7_url": "https://feed7.dev/p/it-s-tokens-all-the-way-down-how-rlms-are-different-kevin-madura-alixpar-1w7pcym",
        "reason": "RLMs provide a complementary way to inspect and partition large inputs programmatically; this Signal focuses the same bounded-delegation principle on ambiguous knowledge retrieval across heterogeneous search tools."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-09-16T15:00:39.000Z",
  "modified_at": "2026-09-16T15:00:39.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/if-we-want-them-to-do-knowledge-work-design-them-as-knowledge-agents-ben-1f64i2y",
    "json": "https://feed7.dev/p/if-we-want-them-to-do-knowledge-work-design-them-as-knowledge-agents-ben-1f64i2y.json",
    "markdown": "https://feed7.dev/p/if-we-want-them-to-do-knowledge-work-design-them-as-knowledge-agents-ben-1f64i2y.md"
  }
}