<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Murad Madi | Portfolio &amp; Blog</title><description>Latest technical case studies and articles.</description><link>https://muradmadi.com/</link><language>en-us</language><item><title>Start With an Email. End With a Hypothesis.</title><link>https://muradmadi.com/blog/outbound-inference-engine</link><guid isPermaLink="true">https://muradmadi.com/blog/outbound-inference-engine</guid><description>You&apos;re handed one email address and nothing else. Enrichment tools will waterfall it into a company, a title, a work history — facts. But facts aren&apos;t why anyone replies. The edge is the step almost everyone skips: an AI layer that turns those facts into a hypothesis about what this specific person actually cares about right now.</description><pubDate>Sat, 08 Aug 2026 00:00:00 GMT</pubDate><category>Martech</category><category>Growth Marketing</category><category>Outbound</category><category>Account Intelligence</category><enclosure url="https://muradmadi.com/og/blog/outbound-inference-engine.png" length="0" type="image/png"/></item><item><title>Uplift, Not Propensity: Causal Targeting for the Post-Cookie Web</title><link>https://muradmadi.com/blog/uplift-vs-propensity-post-cookie</link><guid isPermaLink="true">https://muradmadi.com/blog/uplift-vs-propensity-post-cookie</guid><description>You are probably paying your best customers to do exactly what they were going to do anyway. Propensity scoring — targeting whoever is most likely to buy — is the martech default, and it is quietly wrong. This is a field guide to uplift modeling: finding the people who convert only because you nudged them, demonstrated with a runnable simulation that keeps working even after third-party cookies are gone.</description><pubDate>Sat, 08 Aug 2026 00:00:00 GMT</pubDate><category>Martech</category><category>Growth Marketing</category><category>Causal Inference</category><category>Privacy Sandbox</category><enclosure url="https://muradmadi.com/og/blog/uplift-vs-propensity-post-cookie.png" length="0" type="image/png"/></item><item><title>Case Study: How to Build an Edge-Native Privacy Proxy</title><link>https://muradmadi.com/projects/consentguard</link><guid isPermaLink="true">https://muradmadi.com/projects/consentguard</guid><description>Tired of losing 30% of your marketing data just because a user took three pages to accept a cookie banner? ConsentGuard is a framework-agnostic, edge-native privacy proxy that intercepts network requests, safely buffers payloads, and strictly enforces GDPR, all without breaking your third-party tracking SDKs.</description><pubDate>Mon, 18 May 2026 00:00:00 GMT</pubDate><category>Architecture</category><category>Edge Computing</category><category>Privacy Tech</category><enclosure url="https://muradmadi.com/og/projects/consentguard.png" length="0" type="image/png"/></item><item><title>Case Study: How to Build a Self‑Healing Webhook Pipeline</title><link>https://muradmadi.com/projects/leadmend</link><guid isPermaLink="true">https://muradmadi.com/projects/leadmend</guid><description>LeadMend is a demo service that fixes broken, messy webhook payloads without human intervention. It normalizes missing fields, enriches the data with firmographic details, detects duplicates, and routes leads to the right sales channel, all automatically. For anyone tired of brittle integrations, LeadMend shows how a few lines of governance logic can turn garbage data into actionable, high-confidence leads.</description><pubDate>Wed, 13 May 2026 00:00:00 GMT</pubDate><category>Automated Data Pipelines</category><category>Webhook Validation</category><category>Lead Scoring</category><enclosure url="https://muradmadi.com/og/projects/leadmend.png" length="0" type="image/png"/></item></channel></rss>