<?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>Genai on madeddu.xyz</title><link>https://madeddu.xyz/categories/genai/</link><description>Recent content in Genai on madeddu.xyz</description><generator>Hugo -- 0.145.0</generator><language>en-us</language><lastBuildDate>Sun, 28 Dec 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://madeddu.xyz/categories/genai/index.xml" rel="self" type="application/rss+xml"/><item><title>Building a Second Brain with Strands Agents, AgentCore Memory, and Bedrock Knowledge Bases</title><link>https://madeddu.xyz/posts/aws/strands-agentcore-memory/</link><pubDate>Sun, 28 Dec 2025 00:00:00 +0000</pubDate><guid>https://madeddu.xyz/posts/aws/strands-agentcore-memory/</guid><description>I wanted a personal assistant that knows my markdown notes, remembers ongoing ideas across sessions, and helps me navigate POCs, drafts, and technical decisions. Strands Agents, Bedrock AgentCore Memory, and Bedrock Knowledge Bases turned out to be a very interesting stack for that.</description></item><item><title>Your Own Semantic Search Engine for $2/Month: Amazon S3 Vectors</title><link>https://madeddu.xyz/posts/aws/s3-vectors-personal-search/</link><pubDate>Mon, 08 Dec 2025 00:00:00 +0000</pubDate><guid>https://madeddu.xyz/posts/aws/s3-vectors-personal-search/</guid><description>Amazon S3 Vectors is now generally available. I used it to build a cheap semantic search layer for manuals, notes, and personal docs without standing up OpenSearch or another dedicated vector database.</description></item></channel></rss>