Earlier this month, a service was launched to extract structured data from unstructured sources like text and PDF documents, showcasing the capabilities of large language models in 2024. This service, accessible through a simple front-end, is free for non-production use and aims to help developers build their own applications by demonstrating the potential in this area. It supports various file types, schema formats, and models, allowing users to define extraction schemas in natural language and share them with others. An example use case involves extracting financial data from Uber's Q4 2023 earnings call, showcasing how information from natural language transcripts can be structured for analysis. Additionally, the hosted service utilizes LangServe for its core extraction logic, enabling integration into larger workflows and experimentation with the latest tools.