RAG Me Up Documentation

An education-first documentation

This website documents the RAG Me Up framework built by FutureClub, a Dutch AI agency. Instead of just documenting the framework, this website aims to educate the reader in general AI concepts and, more importantly, setting up production-grade RAG pipelines with all components that are supported by RAG Me Up. Rather than being purely theoretical, this documentation - given that is for the RAG Me Up framework - aims to be practical.

This documentation follows a top-down teaching paradigm where helicopter views and general understanding are being built up first before we dive into different parts of the framework. The goal of doing it this way is to first set a common baseline of what RAG is and how it is approached in RAG Me Up and as such allowing the reader to get a clear understanding of which parts to focus on next instead of being forced to read these docs linearly.

Rationale

This documentation and the RAG Me Up framework are written by Erik Tromp. Most of the materials used in this documentation are part of an elective AI and NLP course for second-year master students at the Jheronimus Academy of Data Science. This documentation is an adopted and extended version of the material used in that course and assumes the same level of thinking.

The reader is assumed to be knowledgeable in the field of Python and have a good grasp on AI to begin with. Deep understand of models and RAG in particular however, are not required as they will be introduced and deepened in this documentation. The rationale for combining the RAG Me Up framework as a running example with an education-first documentation like this, is to have the user always give the ability to actually run everything that is being discussed in this documentation.