/design-rag

Design a Retrieval-Augmented Generation (RAG) pipeline including chunking, embedding, retrieval, and context assembly strategies.

# Design RAG System

Design a production-ready RAG (Retrieval-Augmented Generation) pipeline based on the provided requirements.

## Workflow

### Step 1: Load Required Skills

Load these skills for comprehensive design guidance:

- `rag-architecture` - RAG pipeline patterns
- `model-selection` - Embedding and LLM selection
- `token-budgeting` - Cost estimation

### Step 2: Gather Requirements

If not clear from the description, clarify:

- What documents/content will be indexed?
- What queries will users ask?
- What are the latency requirements?
- What is the expected query volume?
- Are t...