Prashant Rohilla has developed an open-source, local-first web interface, Gemini File Search Manager, to manage Google's Gemini File Search API, a fully managed Retrieval Augmented Generation pipeline. The API simplifies building AI apps by handling chunking, embedding, and storage, with a cost-effective pricing model, charging only a one-time fee of $0.15 per 1 million tokens. The manager, built with Next.js, provides a visual control plane to manage knowledge bases, upload files, and test searches. It supports various file formats and allows users to configure custom chunking and metadata. The interface also features a "Playground" view for conversational history, model selection, and filtering by metadata. The project is MIT licensed and available on GitHub, with plans to add support for structured outputs and persistent chat sessions. This development aims to simplify the use of Gemini File Search API, making it more accessible to developers and scaling applications, and has significant implications for the AI and machine learning industry.