<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Herman Geldenhuys — Essays</title><description>Essays on engineering leadership, AI strategy, and the un-heroic side of leading teams through the AI moment. Written from Montreal.</description><link>https://herman.engineer/</link><language>en-us</language><item><title>Conway&apos;s Law in Reverse</title><link>https://herman.engineer/blog/conways-law-in-reverse/</link><guid isPermaLink="true">https://herman.engineer/blog/conways-law-in-reverse/</guid><description>For nearly sixty years we redrew org charts to fix rigid software. Now the software bends in days, and the org chart has quietly become the most permanent artifact in the building.</description><pubDate>Fri, 31 Jul 2026 00:00:00 GMT</pubDate></item><item><title>The Pin Factory and the Poem</title><link>https://herman.engineer/blog/the-pin-factory-and-the-poem/</link><guid isPermaLink="true">https://herman.engineer/blog/the-pin-factory-and-the-poem/</guid><description>The moment machine intelligence became scarce and metered, our industry began rebuilding the division of labour it thought it had transcended. Adam Smith&apos;s pin factory explains the new agent economy, and his forgotten fine print explains where it has to stop.</description><pubDate>Tue, 28 Jul 2026 00:00:00 GMT</pubDate></item><item><title>The First Software That Can Sell Out</title><link>https://herman.engineer/blog/the-first-software-that-can-sell-out/</link><guid isPermaLink="true">https://herman.engineer/blog/the-first-software-that-can-sell-out/</guid><description>For thirty years, software&apos;s superpower was that the millionth copy cost nothing. In 2026, a frontier AI lab hung a sold-out sign on its door, and the economics of intelligence quietly changed shape.</description><pubDate>Tue, 21 Jul 2026 00:00:00 GMT</pubDate></item><item><title>Worth Doing Badly</title><link>https://herman.engineer/blog/worth-doing-badly/</link><guid isPermaLink="true">https://herman.engineer/blog/worth-doing-badly/</guid><description>Machines just made competence abundant, and everything we ship now arrives looking professional. Which means the most valuable thing left to do with our own hands may be the thing we do badly.</description><pubDate>Wed, 08 Jul 2026 00:00:00 GMT</pubDate></item><item><title>The Field Manual for Agentic Engineering</title><link>https://herman.engineer/blog/the-field-manual-for-agentic-engineering/</link><guid isPermaLink="true">https://herman.engineer/blog/the-field-manual-for-agentic-engineering/</guid><description>OpenAI&apos;s most productive engineering team has not written a line of code in five months. They have shipped about a million lines. They just published what they learned. Here is the executive translation.</description><pubDate>Wed, 06 May 2026 00:00:00 GMT</pubDate></item><item><title>Five Notes from the Sequoia Stage</title><link>https://herman.engineer/blog/karpathy-sequoia-five-notes/</link><guid isPermaLink="true">https://herman.engineer/blog/karpathy-sequoia-five-notes/</guid><description>Andrej Karpathy spent thirty mostly technical minutes on the Sequoia stage describing the executive shift of the decade. Five things from it survive translation.</description><pubDate>Thu, 30 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Why Faster AI Doesn&apos;t Make Faster Companies</title><link>https://herman.engineer/blog/why-faster-ai-doesnt-make-faster-companies/</link><guid isPermaLink="true">https://herman.engineer/blog/why-faster-ai-doesnt-make-faster-companies/</guid><description>AI got an order of magnitude faster. Your company got two or three times faster. The gap is not a mystery. It is a measurement problem.</description><pubDate>Thu, 23 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The Great Inversion</title><link>https://herman.engineer/blog/the-great-inversion/</link><guid isPermaLink="true">https://herman.engineer/blog/the-great-inversion/</guid><description>For 25 years we built interfaces for humans. Now we are building them for agents. The web just flipped.</description><pubDate>Fri, 17 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Building for the Wrong Model</title><link>https://herman.engineer/blog/building-for-the-wrong-model/</link><guid isPermaLink="true">https://herman.engineer/blog/building-for-the-wrong-model/</guid><description>Most enterprise AI investments are compensating for limitations that are about to disappear. When they do, the scaffolding becomes the problem.</description><pubDate>Thu, 16 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Trust, But Verify</title><link>https://herman.engineer/blog/trust-but-verify/</link><guid isPermaLink="true">https://herman.engineer/blog/trust-but-verify/</guid><description>AI is confidently wrong some percentage of the time. The skill is not checking every line. It is knowing which lines to check.</description><pubDate>Fri, 10 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Leeloo Dallas, Memorypass</title><link>https://herman.engineer/blog/me-palace-memory-claude-code/</link><guid isPermaLink="true">https://herman.engineer/blog/me-palace-memory-claude-code/</guid><description>Leeloo learned 5,000 years of human history in ten seconds. Your AI can&apos;t remember last Tuesday. So Milla Jovovich built an open-source fix.</description><pubDate>Wed, 08 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Say What You Mean</title><link>https://herman.engineer/blog/say-what-you-mean/</link><guid isPermaLink="true">https://herman.engineer/blog/say-what-you-mean/</guid><description>Prompting did not die. It fragmented into four different skills. Most of us are still practicing only the first one.</description><pubDate>Wed, 08 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The Taste Gap</title><link>https://herman.engineer/blog/the-taste-gap/</link><guid isPermaLink="true">https://herman.engineer/blog/the-taste-gap/</guid><description>The most productive users of AI are the ones who are best at rejecting its output.</description><pubDate>Sun, 05 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Think in Pieces</title><link>https://herman.engineer/blog/think-in-pieces/</link><guid isPermaLink="true">https://herman.engineer/blog/think-in-pieces/</guid><description>The skill is not solving the problem. It is slicing it. And it applies to far more than code.</description><pubDate>Sat, 04 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The Spec Is the Product</title><link>https://herman.engineer/blog/what-do-you-actually-want/</link><guid isPermaLink="true">https://herman.engineer/blog/what-do-you-actually-want/</guid><description>The most expensive sentence in 2026 is not &apos;build this.&apos; It is &apos;build... something like this, you know what I mean.&apos;</description><pubDate>Fri, 03 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The New Management</title><link>https://herman.engineer/blog/the-new-management/</link><guid isPermaLink="true">https://herman.engineer/blog/the-new-management/</guid><description>AI agents are fast, moderately reliable, and have no sense of direction. The skill they demand most is not technical. It is managerial.</description><pubDate>Wed, 01 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The Hitchhiker&apos;s Guide to the K-Shaped Economy</title><link>https://herman.engineer/blog/hitchhikers-guide-new-job-market/</link><guid isPermaLink="true">https://herman.engineer/blog/hitchhikers-guide-new-job-market/</guid><description>AI changed what we can build. It also changed what we need to know. Here are the six skills that matter now.</description><pubDate>Sat, 28 Mar 2026 00:00:00 GMT</pubDate></item><item><title>It Is Time to Boil the Ocean</title><link>https://herman.engineer/blog/it-is-time-to-boil-the-ocean/</link><guid isPermaLink="true">https://herman.engineer/blog/it-is-time-to-boil-the-ocean/</guid><description>AI is not competition. It is rocket fuel. The question is what we finally aim at.</description><pubDate>Fri, 27 Mar 2026 00:00:00 GMT</pubDate></item><item><title>The Kitchen Is Open</title><link>https://herman.engineer/blog/the-kitchen-is-open/</link><guid isPermaLink="true">https://herman.engineer/blog/the-kitchen-is-open/</guid><description>The kitchen just opened. The question is not who gets cut. It is what we finally get to cook.</description><pubDate>Fri, 20 Mar 2026 00:00:00 GMT</pubDate></item><item><title>I Know Kung Fu: Loading Only What the Mission Needs</title><link>https://herman.engineer/blog/i-know-kung-fu/</link><guid isPermaLink="true">https://herman.engineer/blog/i-know-kung-fu/</guid><description>Claude Code skills are not plugins. They are not tools. They are knowledge, uploaded at the precise moment it&apos;s needed.</description><pubDate>Thu, 19 Feb 2026 00:00:00 GMT</pubDate></item><item><title>The $690B Question Wall Street Can&apos;t Answer</title><link>https://herman.engineer/blog/the-690b-question/</link><guid isPermaLink="true">https://herman.engineer/blog/the-690b-question/</guid><description>Wall Street thinks AI infrastructure is overbuilt and AI is powerful enough to destroy $800B in SaaS market cap. Both cannot be true. The arithmetic picks a winner.</description><pubDate>Tue, 17 Feb 2026 00:00:00 GMT</pubDate></item><item><title>Clawdbought: How OpenAI Acquired the Biggest Open-Source Agent</title><link>https://herman.engineer/blog/clawdbought/</link><guid isPermaLink="true">https://herman.engineer/blog/clawdbought/</guid><description>Two weeks ago, 147,000 stars was a signal. Now the creator of OpenClaw is inside OpenAI. The agent race just changed shape.</description><pubDate>Mon, 16 Feb 2026 00:00:00 GMT</pubDate></item><item><title>When Intelligence Organizes Itself</title><link>https://herman.engineer/blog/when-intelligence-organizes-itself/</link><guid isPermaLink="true">https://herman.engineer/blog/when-intelligence-organizes-itself/</guid><description>AI didn&apos;t copy our org chart. It reinvented it. Three companies, working independently, arrived at the same management structure. That should change how you think about everything.</description><pubDate>Thu, 12 Feb 2026 00:00:00 GMT</pubDate></item><item><title>147,000 GitHub Stars in Two Weeks. Mac Mini Shortages. A 24% Stock Surge. The Market Just Told You Something.</title><link>https://herman.engineer/blog/147000-github-stars/</link><guid isPermaLink="true">https://herman.engineer/blog/147000-github-stars/</guid><description>OpenClaw isn&apos;t the future. But the demand it exposed is.</description><pubDate>Tue, 03 Feb 2026 00:00:00 GMT</pubDate></item><item><title>Claude Has Character. Your AI Has a Script.</title><link>https://herman.engineer/blog/claude-has-character/</link><guid isPermaLink="true">https://herman.engineer/blog/claude-has-character/</guid><description>For thirty years we&apos;ve known that scripts fail under pressure and principles don&apos;t. We applied this to people. We forgot to apply it to machines. Anthropic just remembered.</description><pubDate>Fri, 30 Jan 2026 00:00:00 GMT</pubDate></item><item><title>Anthropic Is Quietly Winning the AI Race</title><link>https://herman.engineer/blog/anthropic-quietly-winning/</link><guid isPermaLink="true">https://herman.engineer/blog/anthropic-quietly-winning/</guid><description>The company most obsessed with what AI should not do has become the company doing the most interesting things with it.</description><pubDate>Fri, 23 Jan 2026 00:00:00 GMT</pubDate></item><item><title>Your IDE is a Dead Pixel</title><link>https://herman.engineer/blog/your-ide-is-a-dead-pixel/</link><guid isPermaLink="true">https://herman.engineer/blog/your-ide-is-a-dead-pixel/</guid><description>The IDE was designed for humans who write code. If agents write the code, the IDE becomes what the light table became to Disney animators.</description><pubDate>Tue, 20 Jan 2026 00:00:00 GMT</pubDate></item><item><title>The Ancient Art of Knowing Less</title><link>https://herman.engineer/blog/the-ancient-art-of-knowing-less/</link><guid isPermaLink="true">https://herman.engineer/blog/the-ancient-art-of-knowing-less/</guid><description>When you create a &apos;QA engineer&apos; agent, you&apos;re not loading specialized capabilities. You&apos;re asking a generalist to forget everything except what matters. The sculptor knew this. Now the engineer must learn it.</description><pubDate>Tue, 13 Jan 2026 00:00:00 GMT</pubDate></item><item><title>The Memento Problem</title><link>https://herman.engineer/blog/the-memento-problem/</link><guid isPermaLink="true">https://herman.engineer/blog/the-memento-problem/</guid><description>Why your AI agents need a system, not a bigger brain</description><pubDate>Fri, 09 Jan 2026 00:00:00 GMT</pubDate></item><item><title>The Bookshelf in the Black Hole</title><link>https://herman.engineer/blog/the-bookshelf-in-the-black-hole/</link><guid isPermaLink="true">https://herman.engineer/blog/the-bookshelf-in-the-black-hole/</guid><description>Why mastery is now the thing that makes you lost</description><pubDate>Wed, 07 Jan 2026 00:00:00 GMT</pubDate></item><item><title>The Genius Who Can&apos;t Say &apos;I Don&apos;t Know&apos;</title><link>https://herman.engineer/blog/the-genius-who-cant-say-i-dont-know/</link><guid isPermaLink="true">https://herman.engineer/blog/the-genius-who-cant-say-i-dont-know/</guid><description>Gemini 3 Flash is fast, cheap, and brilliant. It also doesn&apos;t know what it doesn&apos;t know.</description><pubDate>Thu, 18 Dec 2025 00:00:00 GMT</pubDate></item><item><title>Learning FORTRAN</title><link>https://herman.engineer/blog/learning-fortran/</link><guid isPermaLink="true">https://herman.engineer/blog/learning-fortran/</guid><description>What Dorothy Vaughan understood about the future that most people miss</description><pubDate>Mon, 15 Dec 2025 00:00:00 GMT</pubDate></item><item><title>MCP and the Token Tax</title><link>https://herman.engineer/blog/mcp-and-the-token-tax/</link><guid isPermaLink="true">https://herman.engineer/blog/mcp-and-the-token-tax/</guid><description>Letting your AI agents think</description><pubDate>Mon, 08 Dec 2025 00:00:00 GMT</pubDate></item></channel></rss>