skills.rest
Veloxia-agency
Veloxia-agencyCOMMUNITY
Explore all skills by this author →
GitHub

rag-architect

Design production RAG pipelines with chunking, embeddings, and vector databases.

Updated Apr 24, 2026
Search Tags:#embeddings#rag#evaluation#retrieval#vector search#chunking#reranking
One-click install
npx skills add https://github.com/Veloxia-agency/VELOXIA-WEB --skill rag-architect-veloxia-agency
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rag-architect
Source: https://github.com/Veloxia-agency/VELOXIA-WEB/tree/main/.claude/skills/engineering/skills/rag-architect
Command: npx skills add https://github.com/Veloxia-agency/VELOXIA-WEB --skill rag-architect-veloxia-agency

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you design, tune, and evaluate retrieval-augmented generation systems so your AI answers stay grounded in the right documents instead of guessing.

Core Features & Use Cases

  • Pipeline design: Choose chunking, embeddings, vector databases, retrieval methods, and reranking for a complete RAG architecture.
  • Optimization guidance: Improve retrieval quality with better chunking, hybrid search, query transformation, context assembly, and cost controls.
  • Evaluation and production readiness: Plan metrics, monitoring, caching, fallback behavior, and safety guardrails for reliable deployment.
  • Use case: You have a technical knowledge base and need a production RAG stack that balances accuracy, latency, budget, and maintainability.

Quick Start

Ask the Skill to design a RAG pipeline for your document corpus, query patterns, latency target, budget, and accuracy goals.

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 a technical knowledge base?▼

Designing a RAG pipeline requires selecting appropriate chunking strategies, embedding models, vector databases, and retrieval methods to ground AI answers in your documents. You must balance accuracy, latency, budget, and maintainability for production deployment.

What is the best way to improve retrieval accuracy in RAG systems?▼

Improving retrieval accuracy in RAG systems involves optimizing chunking strategies, applying hybrid search, using query transformation, and implementing reranking. These techniques ensure the system retrieves the most relevant context before generation.

How does reranking and hybrid retrieval work in production RAG architectures?▼

Reranking and hybrid retrieval in RAG architectures combine multiple search signals to refine document relevance before context assembly. This approach filters initial vector search results to prioritize the highest-quality passages for the language model.

How do I evaluate RAG systems for production readiness and safety?▼

Evaluating RAG systems for production readiness requires planning specific metrics, monitoring, caching, and fallback behaviors. You must also implement safety guardrails to ensure reliable deployment and prevent ungrounded or hallucinated outputs.

Can I use this to choose the right embedding models and vector databases?▼

Choosing embedding models and vector databases is a core step of RAG pipeline design. The skill guides selection based on your document corpus, query patterns, and production constraints like latency and cost.

What are the limitations of chunking strategies for document-centric AI systems?▼

Chunking strategies for document-centric AI systems face limitations in balancing context preservation with retrieval precision. Poor chunking can fragment information, requiring hybrid retrieval or query transformation to maintain accuracy within latency and cost constraints.

Related Skills

Explore More Software Engineering Skills→
floflo777floflo777
33

rag-eval

Evaluate RAG systems using retrieval, generation, and latency metrics.

Community
Intermediate
belokonmbelokonm
8

llm-architect

Design and optimize end-to-end LLM systems for production deployment.

Community
Advanced
shalevaminshalevamin

rag-architect

Designs and implements RAG systems with chunking, embedding generation, vector stores, hybrid search, reranking, and evaluation.

Community
Advanced
bcastelinobcastelino
2

rag-engineer

Design and optimize RAG systems for LLM applications.

Community
Advanced