Does it include advanced retrieval techniques like reranking?

Yes, it covers advanced patterns such as Cross-Encoder reranking, hybrid search (combining dense and sparse retrieval), and Maximal Marginal Relevance for result diversity.

Can I use this skill to implement source citations?

Absolutely. It includes prompt engineering templates specifically designed to force the LLM to provide citations and return source document metadata for transparency.

How does this skill handle document chunking?

It includes best practices and code for multiple chunking strategies: recursive character splitting, token-based splitting, semantic chunking, and markdown-header-based splitting.

What is RAG and why is it important for LLM apps?

RAG (Retrieval-Augmented Generation) allows LLMs to access and cite external data not present in their training set, significantly reducing hallucinations and providing context-aware, grounded responses.

Which vector databases does this skill support?

This skill provides implementation patterns and configurations for popular vector databases including Pinecone, Weaviate, Milvus, Chroma, Qdrant, and FAISS.

RAG Implementation Guide

Name: RAG Implementation Guide
Author: nguyendinhquocx

bynguyendinhquocx

0•

Data Science & ML

Builds sophisticated Retrieval-Augmented Generation (RAG) systems to ground LLM responses in external knowledge bases and private documentation.

This skill empowers developers to bridge the gap between static LLMs and dynamic data by implementing robust RAG architectures. It provides comprehensive patterns for document chunking, embedding generation, vector database integration, and advanced retrieval techniques like hybrid search and reranking. Whether you're building a documentation assistant, a research tool, or a domain-specific chatbot, this skill provides the architectural guidance and code snippets necessary to reduce hallucinations and ensure factual accuracy across various LLM frameworks.

Key Features

01Standardized frameworks for evaluating RAG performance using accuracy and groundedness metrics

020 GitHub stars

03Contextual compression and reranking techniques using Cross-Encoders to improve answer relevance

04Seamless integration patterns for leading vector databases like Pinecone, Chroma, and Weaviate

05Advanced document chunking strategies including recursive, semantic, and markdown-aware splitting

06Implementation of complex retrieval patterns such as Hybrid Search and Multi-Query Retrieval

Use Cases

01Building an AI assistant that answers questions based on private company documentation and wikis

02Creating a research tool that provides cited answers from a large corpus of academic papers

03Developing a customer support bot with real-time access to product manuals and technical FAQs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nguyendinhquocx/code-ai rag-implementation

For use in Claude.ai and ChatGPT

Download Skill