<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Built by Matt: Writing</title><description>Notes on what I’m building and learning: decisions, experiments, and what they taught me.</description><link>https://builtbymatt.me/</link><language>en-us</language><item><title>Giving my smartest AI a veto, not a vote</title><link>https://builtbymatt.me/writing/giving-my-smartest-ai-a-veto-not-a-vote/</link><guid isPermaLink="true">https://builtbymatt.me/writing/giving-my-smartest-ai-a-veto-not-a-vote/</guid><description>I expected a smarter AI to fix my wildlife model’s mistakes. Instead it overruled 38 right answers and fixed one. Here’s the job it turned out to be great at.</description><pubDate>Thu, 05 Nov 2026 00:00:00 GMT</pubDate><category>AI</category><category>Architecture</category><category>Projects</category></item><item><title>Teaching an AI to say “none of the above”</title><link>https://builtbymatt.me/writing/teaching-an-ai-to-say-none-of-the-above/</link><guid isPermaLink="true">https://builtbymatt.me/writing/teaching-an-ai-to-say-none-of-the-above/</guid><description>My wildlife app’s model was right 98.9% of the time and still couldn’t admit when it didn’t know. The fix was already sitting in results I’d saved.</description><pubDate>Wed, 04 Nov 2026 00:00:00 GMT</pubDate><category>AI</category><category>Evaluation</category><category>Projects</category></item></channel></rss>