Dottxt, a U.S.-based startup, has raised $11.9 million to improve how AI models understand and generate structured data. Founded by the team behind the open-source project Outlines, Dottxt aims to bridge the gap between traditional software engineering and generative AI workflows. The startup's structured generation approach allows developers to guide large language models more effectively. CEO Rémi Louf emphasizes the importance of grammar and structured text in maximizing AI utility. With growing demand for structured generation tools, Dottxt plans to expand its remote team and focus on enterprise clients in the coming months.