<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Rahim Hirani: Writing</title><description>Notes on engineering leadership, fintech, cloud architecture and AI from a technology leader with 28 years in software.</description><link>https://rahimhirani.com/</link><language>en-gb</language><item><title>Your AI memory problem is a permissions problem</title><link>https://rahimhirani.com/blog/ai-memory-is-a-permissions-problem/</link><guid isPermaLink="true">https://rahimhirani.com/blog/ai-memory-is-a-permissions-problem/</guid><description>Every agent memory pitch is about recall. What stopped me building one company-wide was access control, and the systems that already solve it are the ones you&apos;re already paying for.</description><pubDate>Wed, 21 Oct 2026 00:00:00 GMT</pubDate><category>ai agents</category><category>enterprise ai</category><category>security</category><category>architecture</category></item><item><title>Silence is not health</title><link>https://rahimhirani.com/blog/silence-is-not-health/</link><guid isPermaLink="true">https://rahimhirani.com/blog/silence-is-not-health/</guid><description>The most dangerous output an automated system can produce is a clean one. What I learnt when every scheduled job I relied on died quietly, and the checks I now use to tell &apos;nothing to do&apos; from &apos;nothing worked&apos;.</description><pubDate>Mon, 19 Oct 2026 00:00:00 GMT</pubDate><category>engineering leadership</category><category>reliability</category><category>ai agents</category></item></channel></rss>