---
title: "Discussion on \"Chunking Strategies in RAG: A Deep Dive\" | Hashnode"
description: "Discussion on \"Chunking Strategies in RAG: A Deep Dive\". RAG systems live or die on one decision that happens before any embedding is computed: how the source document gets cut into pieces. This post walks through three chunking mechanisms available in Lang"
url: "https://hashnode.com/posts/chunking-strategies-in-langchain-rag-a-deep-dive/6a4caa488e4035143b326a6b"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Chunking Strategies in RAG: A Deep Dive](https://gauravbytes.dev/chunking-strategies-in-langchain-rag-a-deep-dive?utm_source=hashnode&utm_medium=feed)

RAG systems live or die on one decision that happens before any embedding is computed: how the source document gets cut into pieces. This post walks through three chunking mechanisms available in Lang

Jul 7 · 8 min read · gauravbytes.dev  
Read the article

-   [#agentic-ai](https://hashnode.com/tag/agentic-ai)
-   [#rag](https://hashnode.com/tag/rag)
-   [#ai](https://hashnode.com/tag/ai)
-   [#agents](https://hashnode.com/tag/agents)
-   [#python](https://hashnode.com/tag/python)
-   [#langchain](https://hashnode.com/tag/langchain)
-   [#chunking](https://hashnode.com/tag/chunking)

## Written by

[Gaurav Kumar](https://hashnode.com/@gauravbytes)

@gauravbytes · New Delhi

  
[View profile](https://hashnode.com/@gauravbytes)

Backend Engineer & AI Builder

Backend engineer building ShiftMailer, an AI-powered email tool. Sharing lessons from shipping AI agents and scalable systems. Follow for real-world AI, product, and engineering insights—no fluff.

## Responses 1

Markdown supported · ⌘↵ to post

[Kartik N V J K](https://hashnode.com/@kartik-nvj)

[Jul 10](https://hashnode.com/posts/chunking-strategies-in-langchain-rag-a-deep-dive/6a4caa488e4035143b326a6b/comment/6a511f0634a9f3e0185c825a)

The measurement that only 5% of fixed-size boundaries landed on a sentence terminator is the kind of number more chunking posts should report, because it makes the heading-split failure concrete instead of hand-waved. One thing I have seen bite people with SemanticChunker is that the similarity threshold is corpus-sensitive, so a value tuned on prose quietly over-splits on code-heavy or tabular docs where sentence embeddings are noisy. Did you measure retrieval quality downstream of the three splitters, or just the chunk boundary statistics, since the boundary that reads cleanest to a human is not always the one that gives the best recall?
