skills.rest
zhangzhang-111-i
zhangzhang-111-iCOMMUNITY
Explore all skills by this author →
GitHub

rag-architect

Designs RAG pipelines with chunking, embedding, retrieval, and evaluation strategies.

2|Updated Mar 13, 2026
Search Tags:#llm#rag#vector database#retrieval#pipeline design#embedding
One-click install
npx skills add https://github.com/zhangzhang-111-i/claude-skills111 --skill rag-architect-zhangzhang-111-i
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-architect
Source: https://github.com/zhangzhang-111-i/claude-skills111/tree/main/engineering/rag-architect
Command: npx skills add https://github.com/zhangzhang-111-i/claude-skills111 --skill rag-architect-zhangzhang-111-i

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides comprehensive tools and knowledge for designing, implementing, and optimizing production-grade RAG pipelines, addressing the complexity of building scalable and accurate retrieval systems.

Core Features & Use Cases

  • Document Processing: Offers diverse chunking strategies (fixed-size, sentence, paragraph, semantic) to optimize text segmentation.
  • Model & DB Selection: Guides choices for embedding models and vector databases based on performance and cost.
  • Retrieval & Evaluation: Details advanced retrieval techniques (hybrid, reranking) and robust evaluation frameworks (RAGAS, faithfulness, relevance).
  • Use Case: A developer needs to build a RAG system for customer support documentation. This Skill helps them choose the best chunking method for technical articles, select an appropriate embedding model like all-mpnet-base-v2, recommend a scalable vector database like Qdrant, and set up an evaluation process to ensure answer accuracy and relevance.

Quick Start

Design a RAG pipeline for a technical documentation use case with high accuracy requirements.

Frequently Asked Questions about rag-architect

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

FAQPage Schema
How do I design a RAG pipeline for technical documentation with high accuracy requirements?▼

To design a RAG pipeline, you must select optimal chunking strategies, embedding models, vector databases, and retrieval techniques. For technical documentation, combining semantic chunking with hybrid retrieval ensures scalable and accurate document processing.

What is the best chunking strategy for processing technical articles in a retrieval system?▼

The best chunking strategy for technical articles in retrieval systems is typically semantic chunking. This approach preserves contextual boundaries within the text, ensuring that the embedding model captures accurate meaning for the vector database.

How does hybrid retrieval and reranking improve RAG system performance?▼

Hybrid retrieval and reranking improve RAG performance by combining multiple search methods to recall relevant documents, then reordering them based on semantic relevance. This ensures the language model receives the most accurate context for generation.

Can I use Qdrant as a vector database with an all-mpnet-base-v2 embedding model?▼

Yes, you can use Qdrant as a scalable vector database alongside the all-mpnet-base-v2 embedding model. This combination is recommended for building scalable RAG systems, providing efficient vector storage and accurate document embeddings.

How do I evaluate RAG pipeline accuracy and relevance using frameworks like RAGAS?▼

You evaluate RAG pipeline accuracy using frameworks like RAGAS to measure faithfulness and relevance. Setting up a robust evaluation process ensures the retrieved context aligns with the generated answers, maintaining system reliability across document updates.

When should I not use semantic chunking in a document processing pipeline?▼

Semantic chunking may not be suitable for document processing pipelines dealing with highly structured or uniformly formatted data, where fixed-size chunking is more efficient. It requires computational overhead, so avoid it when processing speed outweighs context preservation.

Related Skills

Explore More Software Engineering Skills→
tranhieutttranhieutt
70

rag-engineer

Designs hybrid semantic and keyword retrieval pipelines for AI knowledge bases.

Community
Advanced
ovachieverovachiever
52

tensorrt-llm

Optimize LLM inference on NVIDIA GPUs with TensorRT-LLM.

Community
Advanced
rukikotoorukikotoo
6

rag

Query the econometrics RAG knowledge base to retrieve relevant document chunks.

Community
Intermediate
automatedigitalautomatedigital
1

unsloth

Fine-tune large language models with LoRA and QLoRA optimization.

Official
Intermediate