<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/"><channel><title>Md. Faruk Alam</title><link>https://farukalamai.com/blog/</link><description>Writing on computer vision, GPU programming, and edge AI — from someone building these systems in production.</description><generator>Hugo</generator><language>en-US</language><copyright>2026 Md. Faruk Alam</copyright><lastBuildDate>Sun, 09 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://farukalamai.com/blog/index.xml" rel="self" type="application/rss+xml"/><item><title>What Your Object Detector Actually Sees in RF-DETR Feature Maps</title><link>https://farukalamai.com/blog/what-your-object-detector-actually-sees-in-rf-detr-feature-maps/</link><guid>https://farukalamai.com/blog/what-your-object-detector-actually-sees-in-rf-detr-feature-maps/</guid><pubDate>Sun, 09 Aug 2026 00:00:00 +0000</pubDate><dc:creator>Md. Faruk Alam</dc:creator><category>Computer Vision</category><category>Deep Learning</category><category>RF-DETR</category><category>Object Detection</category><category>DINOv2</category><description>A detector gives you a box, a class, and a score. None of them tell you why it failed. The feature maps do, and PCA makes them viewable.</description></item><item><title>Before Bangladesh Builds an AI Data Centre, It Needs to Fix Load Shedding</title><link>https://farukalamai.com/blog/before-bangladesh-builds-an-ai-data-centre-it-needs-to-fix-load-shedding/</link><guid>https://farukalamai.com/blog/before-bangladesh-builds-an-ai-data-centre-it-needs-to-fix-load-shedding/</guid><pubDate>Fri, 07 Aug 2026 00:00:00 +0000</pubDate><dc:creator>Md. Faruk Alam</dc:creator><category>Bangladesh</category><category>AI Infrastructure</category><category>Energy</category><category>Climate</category><category>Policy</category><description>There's a $2 billion proposal to build a 200 MW AI data centre in Bangladesh. I work in AI, and I think we should fix the electricity and water first.</description></item><item><title>25 CUDA Concepts Explained, From Threads to Tensor Cores</title><link>https://farukalamai.com/blog/25-cuda-concepts-explained-from-threads-to-tensor-cores/</link><guid>https://farukalamai.com/blog/25-cuda-concepts-explained-from-threads-to-tensor-cores/</guid><pubDate>Wed, 05 Aug 2026 00:00:00 +0000</pubDate><dc:creator>Md. Faruk Alam</dc:creator><category>CUDA</category><category>GPU</category><category>CUDA Concepts</category><category>Parallel Programming</category><category>Accelerated Computing</category><description>A plain-language guide to the 25 CUDA concepts that matter most: kernels, threads, blocks, warps, memory coalescing, shared memory, and streams.</description></item></channel></rss>