---
title: "Comment by Brian · AI News on \"How to build a unified ingestion layer for your RAG pipeline from scratch\" | Hashnode"
description: "Caching embeddings by content hash saved me more than any model swap. Same chunk in, same vector out, and reindexing stops eating the budget."
url: "https://hashnode.com/posts/how-to-build-a-unified-ingestion-layer-for-your-rag-pipeline-from-scratch/6ab3e007dea46a6cc2dce053/comment/6ab48f1e8fc0f59a4c8fad17"
---

[Discussion](https://hashnode.com/posts/how-to-build-a-unified-ingestion-layer-for-your-rag-pipeline-from-scratch/6ab3e007dea46a6cc2dce053)

Comment [How to build a unified ingestion layer for your RAG pipeline from scratch](https://hashnode.com/posts/how-to-build-a-unified-ingestion-layer-for-your-rag-pipeline-from-scratch/6ab3e007dea46a6cc2dce053)

[Brian · AI News](https://hashnode.com/@brianainews)

[Sep 24](https://hashnode.com/posts/how-to-build-a-unified-ingestion-layer-for-your-rag-pipeline-from-scratch/6ab3e007dea46a6cc2dce053/comment/6ab48f1e8fc0f59a4c8fad17)

Caching embeddings by content hash saved me more than any model swap. Same chunk in, same vector out, and reindexing stops eating the budget.

[Het Dave](https://hashnode.com/@enconvert)

[Sep 24](https://hashnode.com/posts/how-to-build-a-unified-ingestion-layer-for-your-rag-pipeline-from-scratch/6ab3e007dea46a6cc2dce053/comment/6ab48f1e8fc0f59a4c8fad17#reply-6ab4ac14c9af724114791a05)

That is indeed true, and is almost always the best method to reduce costs. On a side note, we use this exact same method in one of our other endpoint 'Watch' to detect whether the content of a website has changed or not.