{
  "id": 20620917,
  "title": "Unsiloed nabs $7.5M to turn scanned documents into structured data for AI agents",
  "url": "https://urgent.news/2026/11/18/unsiloed-nabs-7-5m-to-turn-scanned-documents-into-structured-data-for",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-18T16:00:40.000Z",
  "source": {
    "name": "SiliconANGLE",
    "slug": "siliconangle",
    "url": "https://siliconangle.com/2026/11/18/unsiloed-nabs-7-5m-to-turn-scanned-documents-into-structured-data-for-ai-agents/"
  },
  "original_language": "en",
  "account": "Unsiloed Inc., a document-scanning firm, has secured $7.5 million in seed funding to transform scanned documents into structured data for AI agents. The round was led by Felicis and included participation from Y Combinator, Orange Collective, Uncorrelated Ventures, Aarthi Ramamurthy, and Beth Turner.\n\nVarious documents, such as financial statements, insurance claims, contracts, clinical records, and operational forms, often arrive in formats other than text or straight data. These documents, which may include charts, layouts, and other elements, pose challenges in converting them into structured data that can be easily utilized by AI agents.\n\nUnsiloed's technology employs intelligent visual-first scanning to comprehend structured tables, sections, and parent-child relationships between text. The company describes its approach as intelligence-directed and vision-first, enabling it to scan forms, presentations, Word files, and images into Markdown or JSON formats, which are readily readable by large language models.\n\nCEO Aman Mishra believes that AI agents will eventually read and write more documents than humans, necessitating better OCR and a dynamic interface for reading, updating, and acting on unstructured data. Unsiloed goes beyond standard OCR by accounting for page layout and format, preserving original features and relationships between elements.\n\nThe company's parser achieved an 88.0 score on the olmOCR-Bench benchmark, a rigorous evaluation of modern AI models' OCR capabilities. Compared to other services, Unsiloed performed exceptionally well, outperforming Anthropic PBC's Claude Opus 4.7 (81.9), OpenAI GPT 5.5 (84.6), and Google Gemini 3 Pro (77.7) on the benchmark.\n\nUnsiloed's dual-stream vision architecture processes content and visual layouts simultaneously. Developers can upload custom schemas to extract fields, classify, and split documents while preserving formatting. The platform supports private-cloud, on-premises, and fully air-gapped machines, allowing enterprise organizations to integrate Unsiloed into their document pipelines and process data on their own infrastructure.\n\nPrateek Sachan, founder and CTO of Bolna AI, highlights the company's significant improvements in accuracy, context understanding, and response quality across knowledge-base-powered conversations.",
  "summary": "Document-scanning firm Unsiloed Inc. today announced it has raised $7.5 million in seed funding to turn PDFs and other complex documents into structured, readable data for artificial intelligence agents. Participants in the round, led by Felicis, included Y Combinator, Orange Collective, Uncorrelated Ventures, Aarthi Ramamurthy and Beth Turner. A variety of documents arrive in formats […] The…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "TechCrunch",
        "title": "Felicis leads $7.5M bet on Unsiloed AI to make enterprise documents usable by AI agents",
        "url": "https://urgent.news/2026/11/18/felicis-leads-7-5m-bet-on-unsiloed-ai-to-make-enterprise-documents",
        "published": "2026-11-18T12:00:00.000Z"
      }
    ]
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}