<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Lucas Cazelli</title><description>Cognitive science, statistical thinking, and NeuroAI for product management. Building synthetic audiences at North AI.</description><link>https://lucascazelli.com/</link><item><title>The M-System: Multipotentiality Is a Portfolio Problem, Not a Focus Problem</title><link>https://lucascazelli.com/posts/m-system-multipotentiality-portfolio-not-focus/</link><guid isPermaLink="true">https://lucascazelli.com/posts/m-system-multipotentiality-portfolio-not-focus/</guid><description>A Brazilian creator&apos;s video on why we abandon our hobbies turns out to rest on a 1979 change-management model and a well-studied theory of effort. The systems-engineering version, for anyone whose interest list spans robotics, data science, and everything between.</description><pubDate>Wed, 05 Aug 2026 10:00:00 GMT</pubDate></item><item><title>Local or Cloud for Your AI Coding Workload: the 2026 Decision Framework</title><link>https://lucascazelli.com/posts/local-vs-cloud-ai-coding-2026/</link><guid isPermaLink="true">https://lucascazelli.com/posts/local-vs-cloud-ai-coding-2026/</guid><description>Chip export controls and over-capitalised AI competitors are reshaping where teams run inference. A practical framework for choosing between local and cloud AI coding workloads in 2026.</description><pubDate>Thu, 25 Jun 2026 10:00:00 GMT</pubDate></item><item><title>The Labor Illusion Is the Wrong Bet for Creative Testing</title><link>https://lucascazelli.com/posts/labor-illusion-wrong-bet-creative-testing/</link><guid isPermaLink="true">https://lucascazelli.com/posts/labor-illusion-wrong-bet-creative-testing/</guid><description>We trust AI more when it pretends to think slowly. In creative testing that instinct is backwards: the job is a fast, defensible verdict, proven with real attention data, not theatrical delay.</description><pubDate>Thu, 04 Jun 2026 10:00:00 GMT</pubDate></item><item><title>The IBM Selectric: When Brilliant Engineering Loses to Market Change</title><link>https://lucascazelli.com/posts/ibm-selectric-when-brilliant-engineering-loses-to-market-change/</link><guid isPermaLink="true">https://lucascazelli.com/posts/ibm-selectric-when-brilliant-engineering-loses-to-market-change/</guid><description>How IBM&apos;s 1961 typewriter dominated business for 25 years — and why solving the right problem perfectly didn&apos;t save it from the personal computer. A product lesson on market awareness.</description><pubDate>Sat, 09 May 2026 10:00:00 GMT</pubDate></item><item><title>From Finite Elements to Human Elements: An Engineer&apos;s Journey into Decision Science</title><link>https://lucascazelli.com/posts/finite-elements-to-human-elements-engineering-career/</link><guid isPermaLink="true">https://lucascazelli.com/posts/finite-elements-to-human-elements-engineering-career/</guid><description>How transitioning from structural engineering to AI product leadership revealed that understanding human decision-making is more complex—and more rewarding—than any finite element analysis.</description><pubDate>Fri, 12 Sep 2025 10:00:00 GMT</pubDate></item><item><title>Engineering Foundations of Generative AI: From Rules to Probabilities</title><link>https://lucascazelli.com/posts/engineering-roots-of-generative-ai/</link><guid isPermaLink="true">https://lucascazelli.com/posts/engineering-roots-of-generative-ai/</guid><description>A practical history of early generative AI—how rule-based systems evolved into probabilistic models and neural generators, and the engineering trade-offs that shaped modern AI.</description><pubDate>Thu, 11 Sep 2025 10:00:00 GMT</pubDate></item><item><title>Why Your Brain Hates A/B Tests: System 1 vs System 2 in Product Decisions</title><link>https://lucascazelli.com/posts/why-your-brain-hates-ab-tests-system-1-vs-system-2-product-decisions/</link><guid isPermaLink="true">https://lucascazelli.com/posts/why-your-brain-hates-ab-tests-system-1-vs-system-2-product-decisions/</guid><description>Product managers make 100+ micro-decisions daily. Here&apos;s how Kahneman&apos;s dual-process theory explains why we&apos;re so bad at interpreting test results.</description><pubDate>Wed, 10 Sep 2025 10:00:00 GMT</pubDate></item><item><title>Statistical Significance vs Behavioral Significance: A PM&apos;s Dilemma</title><link>https://lucascazelli.com/posts/statistical-significance-vs-behavioral-significance-pm-dilemma/</link><guid isPermaLink="true">https://lucascazelli.com/posts/statistical-significance-vs-behavioral-significance-pm-dilemma/</guid><description>When 95% confidence isn&apos;t enough - understanding the critical difference between statistical significance and behavioral significance in product decisions.</description><pubDate>Fri, 13 Dec 2024 10:00:00 GMT</pubDate></item><item><title>Testing Auto-Concept Highlighting</title><link>https://lucascazelli.com/posts/test-auto-highlighting/</link><guid isPermaLink="true">https://lucascazelli.com/posts/test-auto-highlighting/</guid><description>A test post to verify that concept terms are automatically highlighted in blog post text.</description><pubDate>Thu, 12 Dec 2024 12:00:00 GMT</pubDate></item></channel></rss>