<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Atomic Commit on Haxlys's Blog</title><link>https://hugo-blog-static-site.haxlys.workers.dev/mental_models/atomic-commit/</link><description>Recent content in Atomic Commit on Haxlys's Blog</description><generator>Hugo -- 0.152.2</generator><language>kr-ko</language><lastBuildDate>Thu, 09 Jul 2026 13:51:57 +0900</lastBuildDate><atom:link href="https://hugo-blog-static-site.haxlys.workers.dev/mental_models/atomic-commit/index.xml" rel="self" type="application/rss+xml"/><item><title>그래프 기반 에이전트 메모리: 공유 기억은 어떻게 안전해지는가</title><link>https://hugo-blog-static-site.haxlys.workers.dev/posts/graph-based-agent-memory/</link><pubDate>Thu, 09 Jul 2026 13:51:57 +0900</pubDate><guid>https://hugo-blog-static-site.haxlys.workers.dev/posts/graph-based-agent-memory/</guid><description>System Design Newsletter의 Graph based Agent Memory 글을 바탕으로, 여러 AI 에이전트가 하나의 기억을 안전하게 읽고 쓰려면 왜 그래프, 스키마, 브랜치, 커밋이 필요한지 정리한다.</description></item></channel></rss>