<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Fabio Cingottini</title><description>Notes on software engineering, architecture and running systems in production.</description><link>https://fabiocingottini.com/</link><language>en</language><item><title>Reverse Neutralization: How to Make an LLM Write Like You</title><link>https://fabiocingottini.com/blog/reverse-neutralization/</link><guid isPermaLink="true">https://fabiocingottini.com/blog/reverse-neutralization/</guid><description>Models draft correct, interchangeable emails. The fix is a dataset built backwards: take emails with a voice, let the model neutralize them, then flip the pairs.</description><pubDate>Mon, 07 Sep 2026 00:00:00 GMT</pubDate></item></channel></rss>