Show HN: Epstein's emails reconstructed in a message-style UI (OCR and LLMs) (github.com)

A new Progressive Web App (PWA) has been developed to organize and browse emails from the Jeffrey Epstein estate documents, released by the U.S. House Committee on Oversight and Government Reform. Utilizing advanced optical character recognition (OCR) and a vision-based large language model (LLM) performance, specifically Qwen 2.5 VL 72B, the application converts thousands of email screenshots into a structured, searchable database featuring an iPhone Messages-style interface. The app, optimized for mobile use and capable of offline functionality, presents a user-friendly experience, complete with chronological message threading, contact lists, and search capabilities. This project is significant for the AI/ML community as it demonstrates the practical application of LLMs and OCR in transforming unstructured image data into organized information. Key technical features include the use of a 20-thread implementation for efficient processing, a client-side SQLite database for performance optimization, and thorough data cleaning methods that include duplicate detection and name standardization. Additionally, the project operates entirely in the browser via WebAssembly, making it widely accessible while showcasing the potential for leveraging AI technologies to enhance data interaction and usability.