<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Engineering on Radical Optimist</title><link>https://radoptimist.org/en/tags/engineering/</link><description>Recent content in Engineering on Radical Optimist</description><generator>Hugo</generator><language>en</language><lastBuildDate>Wed, 01 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://radoptimist.org/en/tags/engineering/index.xml" rel="self" type="application/rss+xml"/><item><title>Ticketing Systems Are an Antipattern for Team Collaboration</title><link>https://radoptimist.org/en/post/ticketing-systems-are-an-antipattern-for-team-collaboration/</link><pubDate>Wed, 01 Apr 2026 00:00:00 +0000</pubDate><guid>https://radoptimist.org/en/post/ticketing-systems-are-an-antipattern-for-team-collaboration/</guid><description>&lt;h1 id="the-ticketing-system-user-interface-is-an-antipattern-for-cross-team-collaboration">The ticketing system user interface is an antipattern for cross-team collaboration&lt;/h1>
&lt;p>If your team spends most of its time managing a ticketing system — filing requests, triaging queues, waiting for answers — you have already made your collaboration legible to a machine.&lt;/p>
&lt;p>That is not a metaphor. Ticketing systems only work for the kind of work AI handles well: routine, well-defined, known destination, repeatable process. If your collaboration looks like a queue, it can be automated. And it will be.&lt;/p></description></item><item><title>When Doing Stopped Being Learning</title><link>https://radoptimist.org/en/post/when-doing-stopped-being-learning/</link><pubDate>Wed, 18 Mar 2026 00:00:00 +0000</pubDate><guid>https://radoptimist.org/en/post/when-doing-stopped-being-learning/</guid><description>&lt;blockquote>
&lt;p>&amp;ldquo;What I cannot create, I do not understand.&amp;rdquo; — Richard Feynman&lt;sup id="fnref:1">&lt;a href="#fn:1" class="footnote-ref" role="doc-noteref">1&lt;/a>&lt;/sup>&lt;/p>
&lt;/blockquote>
&lt;p>&lt;em>In the age of AI, every knowledge worker faces the same hidden trade-off: use the tool to produce, or use the tool to understand. You cannot fully do both. How you arbitrage that tension defines what you become.&lt;/em>&lt;/p>
&lt;hr>
&lt;h2 id="i-when-doing-was-learning">I. When doing was learning&lt;/h2>
&lt;p>Before AI, the coupling was tight. To ship software, you had to understand it. There was no shortcut. Writing the code &lt;em>was&lt;/em> the learning. Debugging &lt;em>was&lt;/em> the understanding. The act of production and the act of comprehension were the same act.&lt;/p></description></item><item><title>The Fast Track to Incompetence</title><link>https://radoptimist.org/en/post/ai-peters-principle/</link><pubDate>Fri, 13 Mar 2026 00:00:00 +0000</pubDate><guid>https://radoptimist.org/en/post/ai-peters-principle/</guid><description>&lt;p>&lt;em>AI, the Peter&amp;rsquo;s Principle, and the rise of the Senior Operator&lt;/em>&lt;/p>
&lt;hr>
&lt;h2 id="i-the-principle-accelerated">I. The principle, accelerated&lt;/h2>
&lt;p>In 1969, Laurence J. Peter observed that in any hierarchy, people rise until they reach their level of incompetence — promoted based on past performance until they land in a role their skills can&amp;rsquo;t support. [1] The ceiling was always there. It just took years to find.&lt;/p>
&lt;p>AI has changed the timeline. As a capability amplifier — not a capability builder — it makes you faster and more productive at tasks you already understand. The ceiling stays exactly where it was. The elevator just got faster.&lt;/p></description></item><item><title>Your Pipeline IS Your Culture</title><link>https://radoptimist.org/en/post/cicd-is-your-culture/</link><pubDate>Sun, 01 Mar 2026 00:00:00 +0000</pubDate><guid>https://radoptimist.org/en/post/cicd-is-your-culture/</guid><description>&lt;p>Three questions tell you more about your company culture than any values poster, offsite keynote, or Glassdoor review.&lt;/p>
&lt;p>&lt;strong>Do you trust your developers?&lt;/strong> Can any engineer push to production without waiting for approval? Or do you require mandatory PR reviews, sign-offs, and staging gates — telling your people &amp;ldquo;we hired you, but we don&amp;rsquo;t trust your judgment&amp;rdquo;?&lt;/p>
&lt;p>&lt;strong>Are you actually customer-centric?&lt;/strong> Are developers talking to customers every week and iterating on what they shipped? Or are they building in the dark for months, guessing what users want through layers of gatekeepers, only to discover they were wrong after a massive big-bang release?&lt;/p></description></item><item><title>The Abuse of Tech Debt</title><link>https://radoptimist.org/en/post/the-abuse-of-tech-debt/</link><pubDate>Fri, 17 Jan 2020 10:10:01 +0000</pubDate><guid>https://radoptimist.org/en/post/the-abuse-of-tech-debt/</guid><description>There is nothing unique about tech debt: just change. A tech debt classification and initial root-cause analysis.</description></item><item><title>One Ring to Stitch Them All and in the Land of API Bind Them</title><link>https://radoptimist.org/en/post/one-ring-to-stitch-them-all-and-in-the-land-of-api-bind-them/</link><pubDate>Thu, 26 Dec 2019 06:01:01 +0000</pubDate><guid>https://radoptimist.org/en/post/one-ring-to-stitch-them-all-and-in-the-land-of-api-bind-them/</guid><description>One Ring to Stitch Them All and in the Land of API Bind Them. GraphQL schema stitching for platform or API PM</description></item><item><title>From Product to Platform — Increased Complexity (2/5)</title><link>https://radoptimist.org/en/post/from-product-to-platform-increased-complexity-2-5/</link><pubDate>Mon, 25 Nov 2019 15:01:01 +0000</pubDate><guid>https://radoptimist.org/en/post/from-product-to-platform-increased-complexity-2-5/</guid><description>&lt;p>Journey of a Product manager, from product to platform: complexity!&lt;/p>
&lt;p>Platforms are more complex than products. This post highlight the specific domains where platforms are more complex than products.&lt;/p></description></item></channel></rss>