{
  "schema_version": "1.1",
  "id": "s8:https://www.youtube.com/watch?v=59AA5kIoqjA",
  "slug": "the-search-engine-for-the-agentic-web-will-bryk-exa-1rk74zn",
  "url": "https://feed7.dev/p/the-search-engine-for-the-agentic-web-will-bryk-exa-1rk74zn",
  "title": "The Search Engine for the Agentic Web — Will Bryk, Exa",
  "why_included": "Exa’s search API is shaped around agent constraints: low latency, narrow token extraction, structured results, and per-customer controls rather than a human search-results page.",
  "summary": "Exa says its agent-focused search serves **more than 5,000 companies** and **400,000 developers**. The API can return documents, structured fields, or a compact extraction such as the most relevant 100 tokens from ten results.",
  "practical_implication": "Treat retrieval output as part of the agent interface: tune latency, domains, time windows, excluded page types, and returned token volume for each workflow. Voice agents and research agents should not inherit the same search configuration.",
  "agent_context": "Exa says its agent-focused search serves **more than 5,000 companies** and **400,000 developers**. The API can return documents, structured fields, or a compact extraction such as the most relevant 100 tokens from ten results.\n\nTreat retrieval output as part of the agent interface: tune latency, domains, time windows, excluded page types, and returned token volume for each workflow. Voice agents and research agents should not inherit the same search configuration.\n\nThe scale figures, quality claims, and forecast that AI searches will surpass human searches in 2026 are company assertions without supporting methodology in the talk. More search volume also does not guarantee accurate or trustworthy evidence.",
  "source": {
    "name": "AI Engineer",
    "url": "https://www.youtube.com/watch?v=59AA5kIoqjA",
    "published_at": "2026-09-16T14:00:23.000Z"
  },
  "source_class": "video",
  "content_type": "Video",
  "layer": "tools",
  "domains": [
    "research",
    "data"
  ],
  "topics": [
    "retrieval",
    "tool-use",
    "coding-agents"
  ],
  "verification": {
    "status": "source_linked",
    "label": "Source Linked",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "The scale figures, quality claims, and forecast that AI searches will surpass human searches in 2026 are company assertions without supporting methodology in the talk. More search volume also does not guarantee accurate or trustworthy evidence."
  ],
  "connected_context": {
    "meaning": "This treats search results as a workflow-specific interface rather than a generic context feed. Provider access is only the starting point: agents need different latency, scope, freshness, exclusion, structure, and token budgets according to the task. It therefore strengthens the case for swappable retrieval plumbing while warning that adoption scale and search volume do not establish evidence quality.",
    "corpus_size": 807,
    "generated_at": "2026-09-18T10:06:25.197Z",
    "connections": [
      {
        "title": "Exa web search free through August 31 on AI Gateway and eve",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/changelog/exa-web-search-free-through-august-31-on-ai-gateway-and-eve",
        "feed7_url": "https://feed7.dev/p/exa-web-search-free-through-august-31-on-ai-gateway-and-eve-14bmt41",
        "reason": "The Gateway integration lowers Exa’s setup barrier, while this Signal identifies the configuration and evaluation work that remains after access—quality, latency, filters, returned context, and eventual cost."
      },
      {
        "title": "Tako Search is free on AI Gateway through September 30",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/changelog/tako-search-is-free-on-ai-gateway-through-september-30th",
        "feed7_url": "https://feed7.dev/p/tako-search-is-free-on-ai-gateway-through-september-30th-16drc8y",
        "reason": "Tako provides a second model-independent search path with overlapping controls, reinforcing the implementation consequence that retrieval providers should be swappable and compared per workflow rather than assumed interchangeable."
      },
      {
        "title": "Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation",
        "source_name": "arXiv",
        "source_url": "https://arxiv.org/abs/2607.05382v1",
        "feed7_url": "https://feed7.dev/p/2607-05382v1-1xo10v8",
        "reason": "SearchGen-Bench supplies a caution behind workflow-specific tuning: naive retrieval can add noise, so exposing more search volume or context is not equivalent to selecting useful evidence."
      },
      {
        "title": "Panniantong/Agent-Reach",
        "source_name": "GitHub",
        "source_url": "https://github.com/Panniantong/Agent-Reach",
        "feed7_url": "https://feed7.dev/p/agent-reach-0huqxvi",
        "reason": "Agent Reach abstracts routing and configuration across retrieval backends; this Signal clarifies what that capability layer must vary, including domain, time window, page type, latency, and token volume."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-09-16T14:00:23.000Z",
  "modified_at": "2026-09-16T14:00:23.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/the-search-engine-for-the-agentic-web-will-bryk-exa-1rk74zn",
    "json": "https://feed7.dev/p/the-search-engine-for-the-agentic-web-will-bryk-exa-1rk74zn.json",
    "markdown": "https://feed7.dev/p/the-search-engine-for-the-agentic-web-will-bryk-exa-1rk74zn.md"
  }
}