Skip to content

RAG (Retrieval-Augmented Generation)

A technique that retrieves relevant documents from a knowledge base and feeds them to the model so it answers grounded in your data.

A typical pipeline: split documents into chunks, embed them into a vector database, retrieve the closest chunks for a query, then pass them as context to the LLM. It reduces hallucination and keeps answers up to date without retraining.

Related resources

LlamaIndex Framework Tutorial
A hands-on tutorial for LlamaIndex, the framework designed for building RAG applications that connect LLMs to custom data sources. It covers data ingestion pipelines, indexing strategies, query engines, and agent-based retrieval, with practical examples for PDF, web, and database connectors. Aimed at developers who want to build domain-specific AI assistants grounded in their own documents.
Haystack RAG Framework
Haystack is an open-source framework by deepset for building production-ready LLM applications, specializing in retrieval-augmented generation (RAG) and question-answering pipelines. It offers modular components for document processing, embedding, and retrieval that connect easily with models from OpenAI, Hugging Face, and more.
Qdrant Vector Database
Qdrant is a high-performance open-source vector database and similarity search engine designed for production use. Written in Rust, it powers RAG pipelines, recommendation systems, and semantic search with rich payload filtering, multi-tenancy support, and both self-hosted and managed cloud deployment options.