Discernis Raises $2.5M for AI That Reads Every Document
Fundraise Insider tracks newly funded startups each week and delivers verified sales leads of C-suite contacts straight to your inbox, so you reach the right people while the funding is still fresh. See pricing.
Discernis has raised $2.5 million in a seed round led by Newfund Capital, with Triple Impact Capital, Remarkable Ventures and C2 Ventures also participating.
The New York company was founded in 2024 and builds legal technology for litigation and investigations, where the volume of material has long outpaced anyone’s ability to read it.
Most e-discovery tools sample. They apply search terms, cluster documents, and put a subset in front of reviewers, which works until the thing that matters only becomes visible across the whole record. Discernis reads every document in a case file instead. The company says its system processes more than 50,000 documents an hour and scales further as hardware is added, which is what makes full collection review practical rather than theoretical.
Security shapes the architecture. The software runs on premises or inside a client’s own secure cloud environment, so legal teams can analyse millions of documents without routing privileged material through third party commercial AI APIs. That approach is central to how the company positions itself against tools that depend on outside model providers.
Founder Rob Kwiatkowski came from AWS, where he specialised in reinforcement learning from human feedback. The product is pitched as something closer to an associate that can review and synthesise very large document collections than a keyword tool with a new interface.
The new capital funds product development and a push into large law firms and corporate legal departments.
Vendors selling into newly funded legal technology companies can reach the right buyers through B2B software sales leads tracked as rounds close.