Find your next open-source project

Search by name, purpose, or keyword — or paste a GitHub repository URL
All skills

rag-architect

Design a production RAG pipeline covering chunking, embeddings, vector store, hybrid search, reranking, and evaluation.

When to use it

You already have documents and a retrieval goal but still need to choose chunking, a vector store, and retrieval metrics before building.

For example

You say

Use rag-architect to design a RAG system for our internal support docs in Qdrant, including chunking, hybrid search, reranking, and an evaluation plan.

You get
  • Architecture diagram
  • Vector store trade-off analysis
  • Evaluation plan with thresholds

How it works

  • Includes code checkpoints for chunk metadata, deduplication, and hybrid search results.
  • Recommends measuring precision@k, recall@k, MRR, and NDCG before LLM integration.