# RAG Best Practices — Production Pipeline Guide 2026

> Comprehensive guide to building production RAG pipelines. Covers chunking strategies, embedding models, vector databases, retrieval techniques, evaluation, and common pitfalls with code examples.

## Install

Paste the prompt below into your AI tool:

## Quick Use

The fastest path to a production RAG pipeline:

```python
# 1. Parse documents
from docling.document_converter import DocumentConverter
docs = DocumentConverter().convert("knowledge_base/")

# 2. Chunk intelligently
from langchain.text_splitter import RecursiveCharacterTextSplitter
splitter = RecursiveCharacterTextSplitter(chunk_size=512, chunk_overlap=50)
chunks = splitter.split_documents(docs)

# 3. Embed and store
from langchain_community.vectorstores import Qdrant
vectorstore = Qdrant.from_documents(chunks, embedding=OpenAIEmbeddings())

# 4. Retrieve and generate
retriever = vectorstore.as_retriever(search_kwargs={"k": 5})
context = retriever.invoke("How does authentication work?")
```

---

## Intro

Retrieval-Augmented Generation (RAG) is the dominant architecture for building AI applications that need access to private data — from customer support bots to internal knowledge bases. This guide covers every stage of a production RAG pipeline with code examples, benchmarks, and lessons learned from real deployments. Best for developers building their first RAG system or optimizing an existing one. Works with: any LLM, any vector database.

---

## Pipeline Stages

### 1. Document Parsing

| Tool | Best For | Accuracy |
|------|----------|----------|
| Docling | PDF with tables/figures | Highest |
| Unstructured | Multi-format (15+ types) | High |
| PyPDF | Simple PDFs | Medium |
| BeautifulSoup | HTML/web pages | High |
| Markitdown | Office docs → Markdown | High |

**Rule**: Use Docling for complex PDFs, Unstructured for everything else.

### 2. Chunking Strategies

| Strategy | When to Use | Chunk Size |
|----------|-------------|------------|
| Fixed-size | Simple documents | 512-1024 tokens |
| Recursive | Code and structured text | 512 tokens |
| Semantic | Dense technical content | Variable |
| Document-level | Short documents (<1K tokens) | Full document |
| Sentence-based | FAQ and Q&A content | 3-5 sentences |

**Best practice**: Start with recursive chunking at 512 tokens with 50-token overlap.

```python
splitter = RecursiveCharacterTextSplitter(
    chunk_size=512,
    chunk_overlap=50,
    separators=["\n\n", "\n", ". ", " ", ""]
)
```

### 3. Embedding Models

| Model | Dimensions | Quality | Speed | Cost |
|-------|-----------|---------|-------|------|
| OpenAI text-embedding-3-large | 3072 | Best | Fast | $0.13/M |
| OpenAI text-embedding-3-small | 1536 | Great | Fastest | $0.02/M |
| Cohere embed-v3 | 1024 | Great | Fast | $0.10/M |
| BGE-large-en | 1024 | Good | Medium | Free (local) |
| all-MiniLM-L6 | 384 | OK | Fastest | Free (local) |

**Best practice**: Use `text-embedding-3-small` for most cases. Switch to `large` only if retrieval quality is critical.

### 4. Vector Database Selection

| Database | Hosted | Self-hosted | Best For |
|----------|--------|-------------|----------|
| Qdrant | Yes | Yes | General purpose, filtering |
| Pinecone | Yes | No | Managed, zero ops |
| Turbopuffer | Yes | No | Serverless, auto-scale |
| ChromaDB | No | Yes | Prototyping, local dev |
| pgvector | No | Yes | Already using PostgreSQL |
| Weaviate | Yes | Yes | Multi-modal, GraphQL |

### 5. Retrieval Techniques

| Technique | Improvement | Complexity |
|-----------|-------------|------------|
| Hybrid search (keyword + semantic) | +15-25% | Low |
| Reranking (Cohere, BGE) | +10-20% | Low |
| Query expansion | +5-15% | Medium |
| Parent document retrieval | +10-20% | Medium |
| HyDE (hypothetical doc embedding) | +5-15% | Medium |
| Multi-query retrieval | +10-15% | Low |

**Best practice**: Always use hybrid search + reranking. It is the highest ROI improvement.

### 6. Evaluation

```python
# Use RAGAS for automated evaluation
from ragas import evaluate
from ragas.metrics import faithfulness, answer_relevancy, context_precision

results = evaluate(
    dataset=test_dataset,
    metrics=[faithfulness, answer_relevancy, context_precision]
)
```

### Common Pitfalls

| Pitfall | Solution |
|---------|----------|
| Chunks too large | Reduce to 512 tokens |
| No overlap between chunks | Add 50-100 token overlap |
| Wrong embedding model | Match model to your language |
| No reranking | Add Cohere reranker (+15% accuracy) |
| Ignoring metadata | Filter by date, source, type |
| No evaluation | Use RAGAS or promptfoo |

### FAQ

**Q: What is RAG?**
A: Retrieval-Augmented Generation is an architecture where an LLM retrieves relevant documents from a knowledge base before generating a response, combining the LLMs reasoning with your private data.

**Q: What chunk size should I use?**
A: Start with 512 tokens and 50-token overlap. Adjust based on your document type and retrieval quality metrics.

**Q: Do I need a vector database for RAG?**
A: For production, yes. For prototyping, ChromaDB (in-memory) works. For production, use Qdrant, Pinecone, or pgvector.

---

## Source & Thanks

> Compiled from production RAG deployments, research papers, and community benchmarks.
>
> Related assets on TokRepo: [Docling](https://tokrepo.com), [Qdrant MCP](https://tokrepo.com), [Haystack](https://tokrepo.com), [Turbopuffer MCP](https://tokrepo.com)

---

<!-- ZH -->

## 快速使用

```python
# 解析 → 分块 → 嵌入 → 检索
from docling.document_converter import DocumentConverter
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.vectorstores import Qdrant

docs = DocumentConverter().convert("knowledge_base/")
chunks = RecursiveCharacterTextSplitter(chunk_size=512).split_documents(docs)
vectorstore = Qdrant.from_documents(chunks, embedding=OpenAIEmbeddings())
```

---

## 简介

RAG（检索增强生成）是构建需要访问私有数据的 AI 应用的主流架构。本指南涵盖生产 RAG 管道的每个阶段：文档解析、分块策略、嵌入模型、向量数据库选型、检索技术和评估方法。附代码示例和实战经验。

---

## 来源与感谢

> 综合自生产 RAG 部署、研究论文和社区基准测试。

---
Source: https://tokrepo.com/en/workflows/7ded33e8-464c-4c8f-b3de-6dcf14c0eaf4
Author: Prompt Lab