Unsiloed AI raises $7.5M to turn enterprise documents into agent-ready data
π November 18, 2026 β’ TechCrunch / SiliconANGLE Unsiloed AI raised a $7.5M seed round led by Felicis, with Y Combinator, Orange Collective, Uncorrelated Ventures and angels Aarthi Ramamurthy and Beth Turner participatin
π November 18, 2026 β’ TechCrunch / SiliconANGLE
Unsiloed AI raised a $7.5M seed round led by Felicis, with Y Combinator, Orange Collective, Uncorrelated Ventures and angels Aarthi Ramamurthy and Beth Turner participating. The San Francisco company, founded by IIT Kharagpur classmates Aman Mishra (CEO) and Adnan Abbas (CTO), went through Y Combinator's Fall 2025 batch and runs a team of 7 people. Its pitch: make enterprise documents readable and actionable for AI agents.
The product parses PDFs, scans, images, tables, charts, PPT and DOCX files into structured Markdown and JSON for agents, preserving layout relationships like headings, charts and parent-child sections. Under the hood: dual-stream vision models processing text and image tokens in separate streams, heat map-based chunking that sends more computing power to figures and charts, and a specialized decoder for industry terminology whose meaning changes across business contexts.
A differentiator is verifiability: the software shows where each extracted fact came from in the original document, so businesses can check results instead of trusting AI output blindly. In its own tests, per a technical brief shared by the startup, Unsiloed reports 99.78% accuracy across roughly 84,000 extracted data points from 542 financial documents supplied by an enterprise customer.
On olmOCR-Bench, a public benchmark by Allen Institute for AI, its startup-run tests produced 88.0 versus 66.0 for Reducto's agentic system and 39.9 for Unstructured across 1,403 PDF pages β first among 19 systems per the startup. TechCrunch, however, notes that the results have not been independently verified. The market is crowded: Microsoft and Amazon sell document AI too, and many teams just use general-purpose models. Financial details, model size and training compute are not disclosed.
Adoption is early but real: an unnamed bank uses Unsiloed for document processing, an unnamed law firm deployed it on its own servers for contracts and old case files, and voice AI startup Bolna AI loads knowledge bases for voice agents through it. Pricing starts at one cent per page, per CEO Aman Mishra. βWe believe AI agents will eventually read and write more documents than humans,β Mishra told SiliconANGLE. βThey need a dynamic interface to read, update and act on unstructured data.β
Impact: for teams wiring agents into finance, legal or healthcare workflows, parsing is no longer a hand-built component. Source-linked extraction makes agent output auditable, which matters where a wrong number in a contract is expensive. At that price, piloting on your own document stack is cheap. The round funds researchers, engineers, sales hires and proprietary datasets, per Chip Dispatch and HyperAI. And the benchmark numbers are the startup's own β validate on your own documents first.
Sources:
techcrunch.com/2026/11/18/felicis-leads-7-5m-bet-on-unsiloed-ai-to-make-enterprise-documents-usable-by-ai-agents
siliconangle.com/2026/11/18/unsiloed-nabs-7-5m-to-turn-scanned-documents-into-structured-data-for-ai-agents
chipdispatch.com/articles/unsiloed-ai-raises-7-5m-to-turn-documents-into-agent-ready-data
roundos.ai/funding/companies/unsiloed-ai
x.com/felicis/status/2123097438604947496
Continue the conversation: https://t.me/AIAutomationArchitecture Β· Bluesky
Originally published by Dev.to AI. Aggregated on AIWithGhost for educational purposes β full credit and traffic to the original publisher.