---
title: "Retrieval"
url: "/learn/ai-systems/retrieval"
category: "ai-systems"
---

Retrieval-augmented generation lives or dies on the retrieval half, which is
also the half most write-ups skip past on their way to the prompt. This
subcategory stays there: chunking, indexing, and how to know whether a change
actually helped.

## Posts

- Your RAG Pipeline Is Fine. Your Chunks Are The Problem. (unpublished): Most RAG debugging starts at the prompt and the model, when the failure usually happened earlier — at chunking. This post runs six chunking strategies against the same 400-page corpus and scores each one on whether the right passage actually made it into the context window, with the eval harness and the golden set published alongside the results.