What happened
Unsiloed AI, a San Francisco-based startup that turns complex enterprise documents into structured data for AI agents, has secured $7.5 million in seed funding, TechCrunch reported. Felicis led the round, and Y Combinator, Orange Collective and Uncorrelated Ventures also invested. Angel investors Aarthi Ramamurthy and Beth Turner joined as well.
The software parses tables, charts and complicated layouts, pulls out the information, and links each result to its location in the original file so users can check it. Co-founder and CEO Aman Mishra said the company relies on dual-stream vision models that treat text and image inputs separately. It also uses heat map-based chunking to find key sections, plus a decoder tuned to industry-specific terminology.
Mishra said an unnamed bank uses the technology to process documents. An unnamed law firm runs it on its own servers for contracts and old case files, and voice AI startup Bolna AI uses it to ingest knowledge bases. Pricing begins at one cent per page. The company has seven employees and intends to hire researchers and engineers, expand sales and build proprietary datasets.
Context
Mishra and co-founder Adnan Abbas were classmates at IIT Kharagpur. They entered Y Combinator's Fall 2025 cohort once they had spent time building the technology. The company faces rivals including specialist document-handling startups Reducto and Unstructured, offerings from Microsoft and Amazon, and general-purpose AI models. Mishra argued that large language models are probabilistic and prone to hallucination.
简报
一封邮件,尽览一周科技要闻。
每周精选值得一读的报道。可随时退订。
