Ragie, a new AI company that enables developers to build AI applications connected to their own data with outstanding results in record time. Now available to the general public, Ragie was built by industry veterans Bob Remeika and Mohammed Rafiq. Ragie has raised $5.5M in funding led by Craft Ventures, Saga VC, Chapter One, and Valor.
Most companies rely on LLM providers like OpenAI and Anthropic combined with a technique called RAG (Retrieval Augmented Generation) for their AI applications. RAG leverages a company’s own data to generate more insightful content than can be generated based on what models have been trained on alone. To do this, companies ingest and index their data in a vector database, feed it into a prompt, and generate more thorough and accurate content.
However, the process of building production applications using RAG is very tedious. Developers must:
- Connect and sync multiple data sources including knowledge base applications and cloud file stores
- Extract meaningful data from a variety of file formats and media types like PDFs, Microsoft Office documents and images
- Implement constantly evolving techniques for chunking and retrieval
- Build a scalable data processing pipeline that is resilient and fast
- Avoid hallucinations and ensure that generated content is accurate
- Use open source frameworks that can be time consuming to learn and set up
Also Read: HumanFirst and Infobip Announce a Partnership to Equip Enterprise Teams with Data + Generative AI
Building a homegrown solution like this is time consuming and, despite a team’s best efforts, produces brittle applications.
Ragie solves this by providing a fully managed RAG-as-a-Service platform for developers. Originally developed as a solution for Glue, David Sacks’s new chat application, Ragie implements a robust data ingest pipeline and retrieval API that uses the latest techniques in RAG for chunking, searching, and re-ranking. Through a streamlined developer experience, developers can connect and sync their applications with data in Google Drive, Notion, and Confluence.
In addition to offering the core functionality developers need for RAG, Ragie also offers advanced features such as “Summary Index” to avoid document affinity problems and “Entity Extraction” for extracting structured data from unstructured documents.
Ragie’s straightforward pricing aligns with the way apps are developed, deployed, and scaled. Ragie includes a free tier for developers to get started building their applications, a pro plan for production, and enterprise for scale.
SOURCE: PRNewsWire