<?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>Maroffo</title><link>https://maroffo.github.io/blog/</link><description>Recent content on Maroffo</description><generator>Hugo -- 0.155.3</generator><language>en-us</language><lastBuildDate>Tue, 31 Mar 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://maroffo.github.io/blog/index.xml" rel="self" type="application/rss+xml"/><item><title>Your AI Harness Is Hand-Crafted. That's the Problem.</title><link>https://maroffo.github.io/blog/posts/2026-03-31-your-ai-harness-is-hand-crafted-thats-the-problem/</link><pubDate>Tue, 31 Mar 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-03-31-your-ai-harness-is-hand-crafted-thats-the-problem/</guid><description>Stanford&amp;rsquo;s Meta-Harness paper argues that optimizing the code around the model matters more than optimizing the model itself. I applied it to my Claude Code setup and got the first real numbers: 3,723 tokens always in context, with no evidence they&amp;rsquo;re the right ones.</description></item><item><title>From Cognitive Load to Cognitive Surrender</title><link>https://maroffo.github.io/blog/posts/2026-03-23-from-cognitive-load-to-cognitive-surrender/</link><pubDate>Mon, 23 Mar 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-03-23-from-cognitive-load-to-cognitive-surrender/</guid><description>A Wharton paper gave a formal name to what I described three weeks ago: when AI thinks for you and you let it. The data is worse than I expected.</description></item><item><title>The Apprenticeship Is Dead. Long Live the Indie Senior.</title><link>https://maroffo.github.io/blog/posts/2026-03-15-the-apprenticeship-is-dead-long-live-the-indie-senior/</link><pubDate>Sun, 15 Mar 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-03-15-the-apprenticeship-is-dead-long-live-the-indie-senior/</guid><description>AI creates a demographic time bomb: it increases demand for senior talent while destroying the pipeline that produces it. What replaces corporate apprenticeship is something closer to Hollywood.</description></item><item><title>Everyone Promised Shorter Workweeks. Then Came the 12-Hour Laws.</title><link>https://maroffo.github.io/blog/posts/2026-03-08-everyone-promised-shorter-workweeks-then-came-the-12-hour-laws/</link><pubDate>Sun, 08 Mar 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-03-08-everyone-promised-shorter-workweeks-then-came-the-12-hour-laws/</guid><description>Keynes predicted 15-hour weeks by 2030. Gates says 2-day weeks within a decade. Meanwhile, Germany, Argentina, and Silicon Valley are legislating or normalizing 12-hour days. The paradox has a name, and it&amp;rsquo;s 161 years old.</description></item><item><title>AI Didn't Reduce My Cognitive Load. It Moved It.</title><link>https://maroffo.github.io/blog/posts/2026-03-06-ai-didnt-reduce-my-cognitive-load-it-moved-it/</link><pubDate>Fri, 06 Mar 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-03-06-ai-didnt-reduce-my-cognitive-load-it-moved-it/</guid><description>Everyone says AI makes developers faster. My experience is the opposite: I arrive exhausted not because I wrote more code, but because I evaluated more decisions. The effort didn&amp;rsquo;t shrink. It transformed.</description></item><item><title>AI Is Breaking the Engineering Talent Pipeline</title><link>https://maroffo.github.io/blog/posts/2026-03-04-ai-is-breaking-the-engineering-talent-pipeline/</link><pubDate>Wed, 04 Mar 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-03-04-ai-is-breaking-the-engineering-talent-pipeline/</guid><description>Three fractures in how we learn, hire, and distribute capability.</description></item><item><title>When Your AI's Second Brain Starts Talking Back</title><link>https://maroffo.github.io/blog/posts/2026-02-26-when-your-ais-second-brain-starts-talking-back/</link><pubDate>Thu, 26 Feb 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-02-26-when-your-ais-second-brain-starts-talking-back/</guid><description>Three layers of knowledge integration between Obsidian and Claude Code: from write-only vault to a feedback loop that promotes patterns into skills.</description></item><item><title>The Missing Step: What a Colleague's Hint Taught Me About AI-Driven Planning</title><link>https://maroffo.github.io/blog/posts/2026-02-19-the-missing-step-what-a-colleagues-hint-taught-me-about-ai-driven-planning/</link><pubDate>Thu, 19 Feb 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-02-19-the-missing-step-what-a-colleagues-hint-taught-me-about-ai-driven-planning/</guid><description>Structured requirements, deviation rules, and the art of selective theft: five concepts borrowed from GSD, twenty rejected.</description></item><item><title>When Your AI Skills Library Gets Too Smart for Its Own Context Window</title><link>https://maroffo.github.io/blog/posts/2026-02-12-when-your-ai-skills-library-gets-too-smart-for-its-own-context-window/</link><pubDate>Thu, 12 Feb 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-02-12-when-your-ai-skills-library-gets-too-smart-for-its-own-context-window/</guid><description>From token optimization to an orchestrated development workflow with Claude Code: teaching an AI your patterns is only part of the problem.</description></item><item><title>I Built an Italian Prison Monitor in 9 Days with AI. Here's What Broke.</title><link>https://maroffo.github.io/blog/posts/2026-02-06-i-built-an-italian-prison-monitor-in-9-days-with-ai/</link><pubDate>Fri, 06 Feb 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-02-06-i-built-an-italian-prison-monitor-in-9-days-with-ai/</guid><description>A personal project that started as &amp;ldquo;can I automate an RSS digest?&amp;rdquo; and became a full-stack monitoring platform with narrative memory, semantic search, and AI-generated editorial commentary.</description></item><item><title>How We Cut Our Go API Latency by 75%</title><link>https://maroffo.github.io/blog/posts/2026-01-23-how-we-cut-our-go-api-latency-by-75-percent/</link><pubDate>Fri, 23 Jan 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-01-23-how-we-cut-our-go-api-latency-by-75-percent/</guid><description>A practical guide to modern Go performance patterns we applied to our feed service at Wishew. P50 latency from ~200ms to ~50ms.</description></item><item><title>Measuring Software Performance: What Changed in 3 Years?</title><link>https://maroffo.github.io/blog/posts/2026-01-01-measuring-software-performance-what-changed-in-3-years/</link><pubDate>Thu, 01 Jan 2026 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2026-01-01-measuring-software-performance-what-changed-in-3-years/</guid><description>From the SPACE Era to the AI &amp;amp; DevEx Revolution: a necessary update.</description></item><item><title>From Skills to Shipping: Building with Claude as a Pair Programmer</title><link>https://maroffo.github.io/blog/posts/2025-11-30-from-skills-to-shipping-building-with-claude-as-a-pair-programmer/</link><pubDate>Sun, 30 Nov 2025 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2025-11-30-from-skills-to-shipping-building-with-claude-as-a-pair-programmer/</guid><description>Another chapter in our automation journey at Wishew, where we bridge the gap between project management and development workflows, and discover what happens when your AI pair programmer actually knows your patterns.</description></item><item><title>From Asking Claude to Code to Teaching Claude Our Patterns: Building Modular AI Skills</title><link>https://maroffo.github.io/blog/posts/2025-11-09-from-asking-claude-to-code-to-teaching-claude-our-patterns-building-modular-ai-skills/</link><pubDate>Sun, 09 Nov 2025 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2025-11-09-from-asking-claude-to-code-to-teaching-claude-our-patterns-building-modular-ai-skills/</guid><description>The best way to work with AI isn&amp;rsquo;t to ask it to write code: it&amp;rsquo;s to teach it your patterns first.</description></item><item><title>From Rubber Ducks to Gemini: AI-Powered Code Reviews in GitLab CI</title><link>https://maroffo.github.io/blog/posts/2025-10-18-from-rubber-ducks-to-gemini-ai-powered-code-reviews-in-gitlab-ci/</link><pubDate>Sat, 18 Oct 2025 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2025-10-18-from-rubber-ducks-to-gemini-ai-powered-code-reviews-in-gitlab-ci/</guid><description>A follow-up to our automation journey, where we integrate AI-powered code reviews directly into our CI/CD pipelines, and discover that Gemini is a much better conversation partner than a rubber duck.</description></item><item><title>Building a Smart ECS Deployment Notifier with AWS Lambda, GitLab, and Slack</title><link>https://maroffo.github.io/blog/posts/2025-10-03-building-a-smart-ecs-deployment-notifier-with-aws-lambda-gitlab-and-slack/</link><pubDate>Fri, 03 Oct 2025 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2025-10-03-building-a-smart-ecs-deployment-notifier-with-aws-lambda-gitlab-and-slack/</guid><description>How we built automated Slack notifications for ECS deployments with smart commit filtering, SSM state tracking, and Terraform IaC at Wishew.</description></item><item><title>DeepSeek tra trasparenza e censura</title><link>https://maroffo.github.io/blog/posts/2025-02-01-deepseek-tra-trasparenza-e-censura/</link><pubDate>Sat, 01 Feb 2025 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2025-02-01-deepseek-tra-trasparenza-e-censura/</guid><description>Un test sui diritti umani in Cina</description></item><item><title>Measuring and Improving Performance in Software Development</title><link>https://maroffo.github.io/blog/posts/2023-01-09-measuring-and-improving-performance-in-software-development/</link><pubDate>Mon, 09 Jan 2023 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2023-01-09-measuring-and-improving-performance-in-software-development/</guid><description>How can you tell if a development team is working effectively and meeting its objectives on time? Some insights from scientific research.</description></item><item><title>Misurazione e miglioramento delle performance nello sviluppo software</title><link>https://maroffo.github.io/blog/posts/2023-01-09-misurazione-e-miglioramento-delle-performance-nello-sviluppo-software/</link><pubDate>Mon, 09 Jan 2023 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/posts/2023-01-09-misurazione-e-miglioramento-delle-performance-nello-sviluppo-software/</guid><description>Come si può sapere se un team di sviluppo sta lavorando in modo efficace e se sta raggiungendo i suoi obiettivi in modo tempestivo? Qualche spunto dalla ricerca scientifica.</description></item><item><title>About</title><link>https://maroffo.github.io/blog/about/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/about/</guid><description>About Massimiliano Aroffo</description></item><item><title>Search</title><link>https://maroffo.github.io/blog/search/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://maroffo.github.io/blog/search/</guid><description/></item></channel></rss>