AI Agent News/stories/felicis-leads-7-5m-seed-round-for-unsiloed-ai-s-enterprise-document-tool.md

Infrastructure

Felicis leads $7.5M seed round for Unsiloed AI's enterprise document tool

Unsiloed AI has raised $7.5 million in seed funding led by Felicis to turn complex business documents into structured data that AI agents can use. The startup says its tool shows where each extracted value came from so customers can verify results.

Source: techcrunch.com/2026/11/18/felicis-leads-7-5m-bet-on-unsiloed-ai-to-make-enterprise-documents-usable-by-ai-agents

Unsiloed AI breaks down complex documents, extracts information from tables, charts and other elements, and converts it into structured data for agents. The software also points to the location of each extracted item in the original document so businesses can check the output. Co-founder and CEO Aman Mishra said the company already works with customers in finance, legal services and healthcare, and named Bolna AI as one that uses it to ingest knowledge bases for voice agents.

The startup says its dual-stream vision models process text and images separately, and that it directs more compute to figures, numbers and charts. In a test on 542 financial documents supplied by an enterprise customer, Unsiloed reported 99.78% accuracy across about 84,000 extracted data points. Those results have not been independently verified. On the olmOCR-Bench benchmark, the company reported a score of 88.0, compared with 66.0 for Reducto's agentic system and 39.9 for Unstructured.

The round was led by Felicis, with participation from Y Combinator, Orange Collective and Uncorrelated Ventures, plus angel investors Aarthi Ramamurthy and Beth Turner. Mishra said the company has seven staff and plans to use the money to hire researchers and engineers, expand sales and build proprietary datasets. Pricing starts at one cent per page, he said.

Why it matters if you build agents

Developers building agents that read contracts, financial statements or other tables can evaluate a dedicated parsing layer that returns structured data with source locations attached.

The accuracy and benchmark figures come from the startup's own tests and have not been independently verified, so teams should run their own evaluations on their documents before relying on them.

Unsiloed competes with Reducto, Unstructured, and offerings from Microsoft and Amazon, so the choice of parser is worth comparing on cost and accuracy for the document types an agent handles.

Get stories like this every weekday

One short email with the day's agent tooling news.

Subscribe