{
  "schema_version": "1.1",
  "id": "s8:https://www.youtube.com/watch?v=iJVxxxHM_Oc",
  "slug": "where-rl-will-take-search-maximilian-david-rumpf-sid-ai-1jeb1sw",
  "url": "https://feed7.dev/p/where-rl-will-take-search-maximilian-david-rumpf-sid-ai-1jeb1sw",
  "title": "Where RL Will Take Search — Maximilian-David Rumpf, SID.ai",
  "why_included": "Delegating retrieval to a trained search subagent can keep noisy results out of the main agent’s context while cutting the token, latency, and cost burden of search-heavy workflows.",
  "summary": "SID frames retrieval as a major agent expense: **30–50% of tokens** may be spent searching before the main task begins. Its RL-trained search model iterates against a database, adjusts filters, and returns ranked results.",
  "practical_implication": "For retrieval-heavy agents, isolate search behind a specialist subagent and pass only selected evidence into the main context. SID reports an average drop from roughly two minutes to **about 5 seconds**, alongside **100× lower cost** than frontier-model search.",
  "agent_context": "SID frames retrieval as a major agent expense: **30–50% of tokens** may be spent searching before the main task begins. Its RL-trained search model iterates against a database, adjusts filters, and returns ranked results.\n\nFor retrieval-heavy agents, isolate search behind a specialist subagent and pass only selected evidence into the main context. SID reports an average drop from roughly two minutes to **about 5 seconds**, alongside **100× lower cost** than frontier-model search.\n\nThese are company-reported results across a mixed benchmark suite, with limited task-level detail in the talk. The model still does not match vector-and-reranker latency, and the claim that RL scaling has not reached a ceiling remains unproven.",
  "source": {
    "name": "AI Engineer",
    "url": "https://www.youtube.com/watch?v=iJVxxxHM_Oc",
    "published_at": "2026-09-16T16:30:16.000Z"
  },
  "source_class": "video",
  "content_type": "Video",
  "layer": "agent",
  "domains": [
    "research",
    "data"
  ],
  "topics": [
    "subagents",
    "harness-engineering",
    "retrieval"
  ],
  "verification": {
    "status": "source_linked",
    "label": "Source Linked",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "These are company-reported results across a mixed benchmark suite, with limited task-level detail in the talk. The model still does not match vector-and-reranker latency, and the claim that RL scaling has not reached a ceiling remains unproven."
  ],
  "connected_context": {
    "meaning": "Search can be treated as a bounded learned subsystem rather than consuming the main agent’s context and frontier-model tokens. The reported latency and cost reductions make specialist retrieval a concrete harness optimization, but limited benchmark detail and slower performance than vector-plus-reranker systems leave its advantage workload-dependent rather than established as a general replacement.",
    "corpus_size": 807,
    "generated_at": "2026-09-18T10:06:11.621Z",
    "connections": [
      {
        "title": "If we want them to do Knowledge Work, design them as Knowledge Agents — Benjamin Clavié, Mixedbread",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=O84lhGc1OOI",
        "feed7_url": "https://feed7.dev/p/if-we-want-them-to-do-knowledge-work-design-them-as-knowledge-agents-ben-1f64i2y",
        "reason": "It operationalizes the knowledge-agent recommendation for specialist researchers by assigning iterative retrieval to a dedicated search model and returning only selected evidence."
      },
      {
        "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": "It fits the narrower subagent pattern that survived the failed fixed pipeline: delegate one bounded investigation while retaining end-to-end task ownership in the main agent."
      },
      {
        "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": "Both reduce main-context load by selecting evidence outside the primary model, but RLMs use programmable inspection and optional delegation whereas SID trains a specialist search policy."
      },
      {
        "title": "Build for the Memo, Not the Demo — Shawn Chan, China Resources Holdings",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=tJFjeMBKbIY",
        "feed7_url": "https://feed7.dev/p/build-for-the-memo-not-the-demo-shawn-chan-china-resources-holdings-0i3s3oo",
        "reason": "Passing only selected evidence saves context, but finance-grade use would still require the provenance, uncertainty, and contradiction controls described by the candidate."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-09-16T16:30:16.000Z",
  "modified_at": "2026-09-16T16:30:16.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/where-rl-will-take-search-maximilian-david-rumpf-sid-ai-1jeb1sw",
    "json": "https://feed7.dev/p/where-rl-will-take-search-maximilian-david-rumpf-sid-ai-1jeb1sw.json",
    "markdown": "https://feed7.dev/p/where-rl-will-take-search-maximilian-david-rumpf-sid-ai-1jeb1sw.md"
  }
}