{
  "schema_version": "1.1",
  "id": "s4:https://vercel.com/blog/how-factory-scaled-its-cloud-backend-to-one-billion-monthly-requests-on-vercel",
  "slug": "how-factory-scaled-its-cloud-backend-to-one-billion-monthly-requests-on-1crznka",
  "url": "https://feed7.dev/p/how-factory-scaled-its-cloud-backend-to-one-billion-monthly-requests-on-1crznka",
  "title": "How Factory scaled its cloud backend to one billion monthly requests on Vercel",
  "why_included": "Factory runs API routes, middleware, and webhooks in one Next.js backend at one billion daily requests and 350ms p95. Its agents also deploy internal tools through Vercel APIs under review guardrails.",
  "summary": "Factory says one Next.js backend handles **one billion daily requests** across API routes, middleware, and webhooks at **350ms p95 or below**. It tuned memory limits and function duration rather than creating separate backend infrastructure.",
  "practical_implication": "Agent-heavy teams can keep deployment APIs behind explicit guardrails: Factory's Droids configure, build, and deploy internal tools, while engineers review high-stakes decisions. Observability and route-level WAF rules cover abusive traffic.",
  "agent_context": "Factory says one Next.js backend handles **one billion daily requests** across API routes, middleware, and webhooks at **350ms p95 or below**. It tuned memory limits and function duration rather than creating separate backend infrastructure.\n\nAgent-heavy teams can keep deployment APIs behind explicit guardrails: Factory's Droids configure, build, and deploy internal tools, while engineers review high-stakes decisions. Observability and route-level WAF rules cover abusive traffic.\n\nThis is a vendor case study, not a reproducible benchmark. The material also contains a conflicting quote about tens of millions of daily requests and does not provide workload, cost, or reliability distributions.",
  "source": {
    "name": "Vercel",
    "url": "https://vercel.com/blog/how-factory-scaled-its-cloud-backend-to-one-billion-monthly-requests-on-vercel",
    "published_at": "2026-08-03T04:00:00.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Engineering Post",
  "layer": "infra",
  "domains": [
    "coding",
    "security"
  ],
  "topics": [
    "cloud-agents",
    "observability",
    "sandboxing"
  ],
  "verification": {
    "status": "official_source",
    "label": "Official Source",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "This is a vendor case study, not a reproducible benchmark. The material also contains a conflicting quote about tens of millions of daily requests and does not provide workload, cost, or reliability distributions."
  ],
  "connected_context": {
    "meaning": "This revises the earlier Factory scale claim upward while weakening confidence in the number: the title says monthly, the brief says daily, and the material also retains a tens-of-millions figure. It still reinforces the prior operating pattern—a unified programmable backend with review, observability, and traffic controls—but remains evidence of feasibility, not architectural superiority or a reproducible benchmark.",
    "corpus_size": 353,
    "generated_at": "2026-08-05T10:04:42.227Z",
    "connections": [
      {
        "title": "How Factory scaled its cloud backend to tens of millions of daily requests on Vercel",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/blog/how-factory-scaled-its-cloud-backend-to-tens-of-millions-of-daily-requests",
        "feed7_url": "https://feed7.dev/p/how-factory-scaled-its-cloud-backend-to-tens-of-millions-of-daily-reques-0hhw6dy",
        "reason": "This is the directly preceding account of the same Factory architecture; the new billion-scale claim expands it but conflicts with both that earlier magnitude and figures inside the new material."
      },
      {
        "title": "Agent Runs now available in the Vercel MCP and CLI",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/changelog/agent-runs-vercel-mcp-cli",
        "feed7_url": "https://feed7.dev/p/agent-runs-vercel-mcp-cli-06cfo04",
        "reason": "Factory’s reliance on observability is made more actionable by agent-accessible run traces that expose reasoning, tool calls, and token usage for production debugging."
      },
      {
        "title": "Project-scoped Tokens",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/changelog/project-scoped-tokens",
        "feed7_url": "https://feed7.dev/p/project-scoped-tokens-1emgc6h",
        "reason": "Project-scoped credentials provide a concrete least-privilege control for the deployment APIs that Factory’s agents use, complementing human review and route-level defenses."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-08-03T04:00:00.000Z",
  "modified_at": "2026-08-03T04:00:00.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/how-factory-scaled-its-cloud-backend-to-one-billion-monthly-requests-on-1crznka",
    "json": "https://feed7.dev/p/how-factory-scaled-its-cloud-backend-to-one-billion-monthly-requests-on-1crznka.json",
    "markdown": "https://feed7.dev/p/how-factory-scaled-its-cloud-backend-to-one-billion-monthly-requests-on-1crznka.md"
  }
}