What is the RAG Architect skill?

It is a specialized capability for Claude Code that provides expert guidance on designing, implementing, and optimizing Retrieval-Augmented Generation systems.

Which vector databases does this skill support?

The skill offers comparative design advice and implementation patterns for popular stores including Pinecone, Weaviate, Chroma, Qdrant, Milvus, and pgvector.

Can it help with evaluating RAG performance?

Yes, it includes frameworks for tracking metrics like precision@k, recall, and MRR, as well as integrating evaluation tools like RAGAS and TruLens.

How does it improve retrieval quality?

It provides strategies for hybrid search, query expansion, semantic reranking, and metadata filtering to ensure the most relevant context is retrieved for the LLM.

RAG Architect

Name: RAG Architect
Author: ngxtm

byngxtm

•

Data Science & ML

Architects and optimizes Retrieval-Augmented Generation (RAG) systems to ground AI applications in factual, domain-specific knowledge.

The RAG Architect skill provides comprehensive guidance for building production-grade retrieval systems, from initial vector database selection and schema design to advanced retrieval strategies like hybrid search and reranking. It assists developers in implementing efficient document ingestion pipelines with smart chunking, selecting optimal embedding models, and establishing robust evaluation frameworks to ensure high-quality, grounded LLM responses while balancing latency, accuracy, and infrastructure costs.

Key Features

01Establish RAG evaluation frameworks using metrics like precision@k, MRR, and RAGAS

02Optimize retrieval pipelines using hybrid search, query expansion, and semantic reranking

03Design scalable vector database architectures including Pinecone, Weaviate, and pgvector

04Implement intelligent document chunking and metadata enrichment strategies

052 GitHub stars

06Generate comprehensive system architecture diagrams for ingestion and retrieval flows

Use Cases

01Building a corporate knowledge base chatbot requiring high factual accuracy

02Integrating private domain documentation into a customer support AI assistant

03Optimizing an existing semantic search system for better relevance and lower latency

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ngxtm/devkit rag-architect

For use in Claude.ai and ChatGPT