<?xml version="1.0" encoding="UTF-8" ?>
<rss version="2.0">
    <channel>
      <title>Yet Another AI Blog</title>
      <link>https://http://www.dlai.dev</link>
      <description>Last 10 notes on Yet Another AI Blog</description>
      <generator>Quartz -- quartz.jzhao.xyz</generator>
      <item>
    <title>About Me</title>
    <link>https://http://www.dlai.dev/</link>
    <guid>https://http://www.dlai.dev/</guid>
    <description>AI Engineering and Inference Optimization Specialist</description>
    <pubDate>Tue, 26 May 2026 16:06:19 GMT</pubDate>
  </item><item>
    <title>Interactive Tools</title>
    <link>https://http://www.dlai.dev/tools/</link>
    <guid>https://http://www.dlai.dev/tools/</guid>
    <description>Hands-on tools for understanding GPU programming and AI optimization</description>
    <pubDate>Tue, 26 May 2026 16:06:19 GMT</pubDate>
  </item><item>
    <title>Hermes Research with khoj-cli</title>
    <link>https://http://www.dlai.dev/2026-05-26-khoj-cli-hermes-research-reflex</link>
    <guid>https://http://www.dlai.dev/2026-05-26-khoj-cli-hermes-research-reflex</guid>
    <description>How a small Exa-powered terminal search tool becomes much more useful when wrapped as a Hermes skill.</description>
    <pubDate>Tue, 26 May 2026 00:00:00 GMT</pubDate>
  </item><item>
    <title>Qwen3.5 on Dual R9700s</title>
    <link>https://http://www.dlai.dev/2026-04-09-qwen35a3b-r9700-rocm-vllm</link>
    <guid>https://http://www.dlai.dev/2026-04-09-qwen35a3b-r9700-rocm-vllm</guid>
    <description>Benchmark notes from dual-R9700 ROCm inference: good warm throughput, decent TTFT, and a tensor-parallel path that still likes to explode under stress.</description>
    <pubDate>Thu, 09 Apr 2026 00:00:00 GMT</pubDate>
  </item><item>
    <title>LLM-Guided Kernel Optimization</title>
    <link>https://http://www.dlai.dev/2025-12-20-llm-kernel-optimization</link>
    <guid>https://http://www.dlai.dev/2025-12-20-llm-kernel-optimization</guid>
    <description>A technical discussion of how large language models can speed up GPU kernel optimization, from research ideas to production kernels.</description>
    <pubDate>Sat, 20 Dec 2025 00:00:00 GMT</pubDate>
  </item><item>
    <title>Triton Linear Layouts</title>
    <link>https://http://www.dlai.dev/2025-06-22-linear-layouts</link>
    <guid>https://http://www.dlai.dev/2025-06-22-linear-layouts</guid>
    <description>Exploring the concept of linear layouts in Triton, a programming language for writing high-performance GPU code, and how it simplifies memory management and data access patterns.</description>
    <pubDate>Sun, 22 Jun 2025 00:00:00 GMT</pubDate>
  </item><item>
    <title>PyTorch 2.x Backends</title>
    <link>https://http://www.dlai.dev/2025-05-26-torch-inductor</link>
    <guid>https://http://www.dlai.dev/2025-05-26-torch-inductor</guid>
    <description>PyTorch 2.x Backends The Great ML Framework Debate Centuries ago (2020), I contributed to PyTorch library, specifically, TorchServe which used to be the default model serving library for PyTorch Models.</description>
    <pubDate>Mon, 26 May 2025 00:00:00 GMT</pubDate>
  </item><item>
    <title>CUDA Threads and Blocks</title>
    <link>https://http://www.dlai.dev/2025-05-17-cuda-basics</link>
    <guid>https://http://www.dlai.dev/2025-05-17-cuda-basics</guid>
    <description>An in-depth exploration of CUDA configurations and their impact on thread/block patterns for optimal performance</description>
    <pubDate>Sat, 17 May 2025 00:00:00 GMT</pubDate>
  </item><item>
    <title>CUDA Thread/Block Visualizer</title>
    <link>https://http://www.dlai.dev/2025-05-17-cuda-visualizer</link>
    <guid>https://http://www.dlai.dev/2025-05-17-cuda-visualizer</guid>
    <description>CUDA Thread/Block Visualizer This interactive tool helps you understand how different thread and block configurations affect GPU execution patterns.</description>
    <pubDate>Sat, 17 May 2025 00:00:00 GMT</pubDate>
  </item><item>
    <title>CUDA Thread/Block Visualizer</title>
    <link>https://http://www.dlai.dev/tools/cuda-visualizer</link>
    <guid>https://http://www.dlai.dev/tools/cuda-visualizer</guid>
    <description>CUDA Thread/Block Visualizer This interactive tool helps you understand how different thread and block configurations affect GPU execution patterns.</description>
    <pubDate>Sat, 17 May 2025 00:00:00 GMT</pubDate>
  </item>
    </channel>
  </rss>