<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Ahmed Mohammed — AI/ML Blog</title>
    <link>https://ahmed-3m.github.io/blog/</link>
    <description>Articles on computer vision, diffusion models, OOD detection, and production AI engineering.</description>
    <language>en-us</language>
    <atom:link href="https://ahmed-3m.github.io/feed.xml" rel="self" type="application/rss+xml"/>
    <managingEditor>ahmed.mo.0595@gmail.com (Ahmed Mohammed)</managingEditor>
    <lastBuildDate>Sat, 25 Apr 2026 19:17:14 GMT</lastBuildDate>
    
    <item>
      <title><![CDATA[How I Hit 99.03% AUROC on OOD Detection Using Conditional Diffusion Models]]></title>
      <link>https://ahmed-3m.github.io/blog/ood-diffusion-thesis/</link>
      <guid isPermaLink="true">https://ahmed-3m.github.io/blog/ood-diffusion-thesis/</guid>
      <description><![CDATA[My Master's thesis at JKU Linz (supervised by Prof. Sepp Hochreiter) introduced class-conditional separation loss into conditional diffusion models used as generative classifiers — improving out-of-distribution detection from 80.25% to 99.03% AUROC on CIFAR-10, an 18.8 percentage-point gain.]]></description>
      <pubDate>Fri, 20 Mar 2026 00:00:00 GMT</pubDate>
      <category>Diffusion Models</category>
      <category>OOD Detection</category>
      <category>Deep Learning</category>
      <category>PyTorch</category>
      <category>CIFAR-10</category>
      <category>Generative Models</category>
      <category>JKU Linz</category>
    </item>
  
    <item>
      <title><![CDATA[Shipping Faultrix: What I Learned Building an AI SaaS From Zero in 5 Months]]></title>
      <link>https://ahmed-3m.github.io/blog/5-month-llm-adventure/</link>
      <guid isPermaLink="true">https://ahmed-3m.github.io/blog/5-month-llm-adventure/</guid>
      <description><![CDATA[I went from a research background (diffusion models, computer vision) to shipping a production AI SaaS — Faultrix, an AI-powered construction quality control platform generating ÖNORM-compliant reports in under 1 minute. Here's what the gap between "AI research" and "AI product" actually looks like.]]></description>
      <pubDate>Sat, 15 Nov 2025 00:00:00 GMT</pubDate>
      <category>LLM</category>
      <category>SaaS</category>
      <category>Next.js</category>
      <category>OpenAI</category>
      <category>Production AI</category>
      <category>Faultrix</category>
      <category>Entrepreneurship</category>
    </item>
  
    <item>
      <title><![CDATA[Diffusion Models for Industrial Defect Detection: 98.4% Accuracy at PROFACTOR GmbH]]></title>
      <link>https://ahmed-3m.github.io/blog/diffusion-models-anomaly-detection/</link>
      <guid isPermaLink="true">https://ahmed-3m.github.io/blog/diffusion-models-anomaly-detection/</guid>
      <description><![CDATA[How I combined YOLOv8 feature extraction with conditional diffusion models to achieve 98.4% defect detection accuracy on inkjet-printed building components in a real-time production environment — part of the Zer0P project funded by the Government of Upper Austria.]]></description>
      <pubDate>Wed, 20 Nov 2024 00:00:00 GMT</pubDate>
      <category>Diffusion Models</category>
      <category>Anomaly Detection</category>
      <category>YOLOv8</category>
      <category>Industrial AI</category>
      <category>Computer Vision</category>
      <category>Quality Control</category>
    </item>
  
    <item>
      <title><![CDATA[YOLOv8 for Industrial Quality Control: Architecture Decisions That Moved the Needle]]></title>
      <link>https://ahmed-3m.github.io/blog/computer-vision-yolo-mastery/</link>
      <guid isPermaLink="true">https://ahmed-3m.github.io/blog/computer-vision-yolo-mastery/</guid>
      <description><![CDATA[The specific architecture choices, data engineering decisions, and production deployment tricks that contributed to 98.4% defect detection accuracy on inkjet-printed building components at PROFACTOR GmbH — not a tutorial, a post-mortem.]]></description>
      <pubDate>Sun, 20 Oct 2024 00:00:00 GMT</pubDate>
      <category>Computer Vision</category>
      <category>YOLOv8</category>
      <category>Deep Learning</category>
      <category>Industrial AI</category>
      <category>PyTorch</category>
      <category>Production ML</category>
    </item>
  
  </channel>
</rss>