Low-Code Modular RAG-Based Knowledge Search Engine

Introduction

How can you enable enterprise-grade retrieval-augmented generation (RAG) applications? An Oracle low-code modular large language model (LLM) search engine can help.

In this solution, you’ll learn how to use Oracle Integration Cloud to integrate and orchestrate social media, productivity, and business app sources and channels; LLMs; and vector databases to ingest live data into the RAG-based knowledge search engine store.

Use Oracle Cloud Infrastructure (OCI) Document Understanding to extract information from different document types. Leverage OCI Generative AI for document summarization, generation, and synthesis of answers. Use Oracle Base Database Service for document extraction, vector search, and embedding into the database using Open Neural Network Exchange local models and apply local OCI Data Science models for better answers from advanced RAG.

Demo

Prerequisites and setup

1. Oracle Cloud account—sign-up page
2. OCI Generative AI—documentation
3. OCI Document Understanding—documentation
4. Oracle Integration Cloud with Oracle Visual Builder—documentation
5. OCI Object Storage—documentation
6. Oracle Base Database Service (23ai)—documentation
7. OCI Data Science—documentation

Getting started

• Detailed steps and sample code on GitHub