<?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>Knowledge-Graph on Ashwin Viswamithiran</title><link>https://ashwinviswamithiran.com/tags/knowledge-graph/</link><description>Recent content in Knowledge-Graph on Ashwin Viswamithiran</description><image><title>Ashwin Viswamithiran</title><url>https://ashwinviswamithiran.com/og-default.png</url><link>https://ashwinviswamithiran.com/og-default.png</link></image><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 30 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://ashwinviswamithiran.com/tags/knowledge-graph/index.xml" rel="self" type="application/rss+xml"/><item><title>It Reads Better Than It Decides</title><link>https://ashwinviswamithiran.com/writing/it-reads-better-than-it-decides/</link><pubDate>Wed, 30 Sep 2026 00:00:00 +0000</pubDate><guid>https://ashwinviswamithiran.com/writing/it-reads-better-than-it-decides/</guid><description>Building an OpenCTI enrichment connector on Gemini: where the model was genuinely good, where it would have done real damage, and the six times running it made me move a decision out of the model and into deterministic code.</description></item><item><title>Teaching a Chatbot to Read a Spider Web</title><link>https://ashwinviswamithiran.com/writing/teaching-a-chatbot-to-read-a-spider-web/</link><pubDate>Thu, 09 Apr 2026 00:00:00 +0000</pubDate><guid>https://ashwinviswamithiran.com/writing/teaching-a-chatbot-to-read-a-spider-web/</guid><description>Why a capable AI assistant could not answer an easy question about a well-documented threat actor, and the unglamorous fix: flattening a threat-intelligence graph into one readable card per entity, ahead of time.</description></item></channel></rss>