We are a group of tech enthusiasts specializing in consulting and product development within the AI, Cloud, and DevOps realms. Over the past two years, as Generative AI (GenAI) started gaining popularity, we began creating GenAI-based solutions for our own products and customers.
One common observation was that everyone was building a RAG (retrieval-augmented generation) application, either from a knowledge base or a database. After our first project, we decided to develop this process into a framework, making it easier for future use. While building a RAG is straightforward for engineers, its maintenance, monitoring, and enhancement require additional resources.
Many small organizations and individuals lack the resources to invest in building their own RAG. That's where we come in, inspired by the magical Genie from Disney who could create things with a snap. Hence, we've developed RAGGENIE, a tool that lets you build your own co-pilot application in an instant.
For those unfamiliar with the terms:
Consider yourself as someone wanting to build a small GenAI-based chat application for personal use or your business. If you don't want to spend much time on technical aspects and prefer quick, finished results, RAGGENIE is for you. It allows you to connect multiple data sources and tools and helps you create a conversational chat interface quickly. Once created, you can chat with your data on our portal, share it as a link with your colleagues, or embed it on your website or portal.
Our initial roadmap is to build a simple RAG application using a few specific sources and following the best practices. The first release will consider security and customization. With it, you'll be able to connect with a few document shares and databases and start your chat. For now, only one source will be supported at a time.
In the first release, you'll need to provide your own InferenceAPI key and Vector store key to get started. Soon, we'll offer the option to use our own subscriptions for these.
These are sources where you might store your application data. If you want to create a GenAI chatbot to analyze your data or integrate it with your product, these sources will be useful. They act as a communicator where RAG creates a query to this platform, queries your actual data, and provides the result. The real data is not indexed to any other system; it comes freshly from your sources.
These sources allow you to use files such as text or Word documents (initial versions do not support PDF) to create an AI chat application that can interact with this data.
We use the following stack for this application, but the underlying RAG application relies on our system.
Supported Inference APIs:
Use RAGGENIE to swiftly build a GenAI-based chat application for personal or business use, connecting multiple data sources to create a conversational interface that can be shared or embedded as desired
You can create your own conversational chat feature for your Application. Then, integrate it into your application as a chat app or embbedd inside application.
Create several chatbots, each tailored to different teams like sales, marketing, and tech. Each chatbot should draw from a different knowledge base and be tuned for different use cases.
Please let us know if you have any feedback for us :
