<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Rag on jonam'Log</title><link>https://www.jonam.io/tags/rag/</link><description>Recent content in Rag on jonam'Log</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>&amp;copy; 2026 Manoj. All Rights Reserved.</copyright><lastBuildDate>Mon, 18 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://www.jonam.io/tags/rag/index.xml" rel="self" type="application/rss+xml"/><item><title>DocVault</title><link>https://www.jonam.io/journal/inference-engineering/product-ideas/docvault/</link><pubDate>Mon, 18 May 2026 00:00:00 +0000</pubDate><guid>https://www.jonam.io/journal/inference-engineering/product-ideas/docvault/</guid><description>Compute any document&amp;rsquo;s context once, serve it to every user forever.</description></item><item><title>Position-Invariant Document KV Cache</title><link>https://www.jonam.io/journal/inference-engineering/research-topics/position-invariant-document-kv-cache/</link><pubDate>Mon, 18 May 2026 00:00:00 +0000</pubDate><guid>https://www.jonam.io/journal/inference-engineering/research-topics/position-invariant-document-kv-cache/</guid><description>Can document KV states be cached independent of prompt position and reused across RAG queries?</description></item><item><title>PrefillX</title><link>https://www.jonam.io/journal/inference-engineering/product-ideas/prefillx/</link><pubDate>Mon, 18 May 2026 00:00:00 +0000</pubDate><guid>https://www.jonam.io/journal/inference-engineering/product-ideas/prefillx/</guid><description>Cut TTFT for long-context document applications by precomputing and repairing reusable KV states.</description></item></channel></rss>