rag-pipeline-architect

Name: rag-pipeline-architect
Availability: InStock
Author: scheune10-coding

Design end-to-end RAG pipelines with vector DB and chunking strategies.

Updated Mar 30, 2026

Search Tags:#architecture #production #embeddings #rag #vector-database #hybrid-retrieval #rag-pipeline

Ready to run

npx skills add https://github.com/Scheune10-coding/my-claude-code-skills --skill rag-pipeline-architect-scheune10-coding

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: rag-pipeline-architect
Source: https://github.com/Scheune10-coding/my-claude-code-skills/tree/main/rag-pipeline-architect
Command: npx skills add https://github.com/Scheune10-coding/my-claude-code-skills --skill rag-pipeline-architect-scheune10-coding

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a proven blueprint for designing end-to-end RAG pipelines, addressing the complexity of selecting suitable vector databases, defining chunking strategies, integrating embeddings, and establishing production-ready deployment patterns.

Core Features & Use Cases

Vector DB selection guidance (Qdrant, pgvector, FAISS, Weaviate, Neo4j) and trade-offs based on scale and requirements.
Chunking strategy guidance by document type (PDF/HTML/Markdown/Code) and use case (Q&A, summarization, chat).
Hybrid retrieval patterns (vector search plus knowledge graph) and semantic caching to reduce LLM usage.
Ingestion design, embedding model selection, and architecture decision outputs for production runtimes.
Production deployment patterns, monitoring, and evaluation considerations.

Quick Start

Define your use case, select a vector DB, and outline ingestion, chunking, and retrieval steps.

Frequently Asked Questions about rag-pipeline-architect

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I design a production-ready RAG pipeline for document QA?▼

Design a production-ready RAG pipeline by selecting a vector database, defining chunking strategies for your document types, integrating embedding models, and establishing semantic caching and monitoring patterns.

What is the best way to choose a vector database for my RAG architecture?▼

Choose a vector database like Qdrant, pgvector, FAISS, Weaviate, or Neo4j by evaluating trade-offs based on your specific scale, retrieval requirements, and production deployment context.

How do I implement hybrid retrieval with knowledge graphs in a RAG system?▼

Implement hybrid retrieval by combining vector search with knowledge graph connections and applying semantic caching to reduce LLM usage and improve retrieval accuracy.

What chunking strategy should I use for PDF and HTML documents in a RAG pipeline?▼

Chunking strategies should be defined by document type, such as PDF, HTML, Markdown, or code, and tailored to your specific use case like question answering, summarization, or chat.

How do I reduce LLM usage costs in a production RAG deployment?▼

Reduce LLM usage costs in a RAG deployment by implementing semantic caching patterns alongside hybrid retrieval and optimized ingestion pipeline designs to avoid redundant model calls.

Does this RAG architecture blueprint support chatbot and search use cases?▼

This RAG architecture blueprint specifies production deployment patterns across document QA, chatbot, and search use cases, providing recommended tech stacks and production-readiness guidelines for each.