<?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>Lead-Triage on Local First AI</title><link>https://localfirstai.eu/tags/lead-triage/</link><description>Recent content in Lead-Triage on Local First AI</description><image><title>Local First AI</title><url>https://localfirstai.eu/images/og-default.png</url><link>https://localfirstai.eu/images/og-default.png</link></image><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 19 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://localfirstai.eu/tags/lead-triage/index.xml" rel="self" type="application/rss+xml"/><item><title>Sorting Estate-Agency Enquiries on a Mac Mini: 89 % Right on 235 Synthetic Messages</title><link>https://localfirstai.eu/posts/2026-10-19-lead-triage/</link><pubDate>Mon, 19 Oct 2026 00:00:00 +0000</pubDate><guid>https://localfirstai.eu/posts/2026-10-19-lead-triage/</guid><description>One small local decision model answers closed questions about an estate-agency enquiry; the agency&amp;#39;s written rules, in code, pick the urgency and the next step. On 235 synthetic enquiries in seven languages it read the request right 89 % of the time, and the rules picked the right next step 90 % of the time. Three genuine enquiries were flagged as spam, so spam now goes to a review folder and is never deleted.</description></item></channel></rss>