<?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>Production on Zhanyl Abdybaeva</title>
    <link>https://zhanyl-tech.github.io/tags/production/</link>
    <description>Recent content in Production on Zhanyl Abdybaeva</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Thu, 15 Jan 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://zhanyl-tech.github.io/tags/production/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Agentic ML Inference Infrastructure</title>
      <link>https://zhanyl-tech.github.io/projects/agentic-ml-inference-infrastructure/</link>
      <pubDate>Thu, 01 Jan 2026 00:00:00 +0000</pubDate>
      <guid>https://zhanyl-tech.github.io/projects/agentic-ml-inference-infrastructure/</guid>
      <description>Production agentic AI infrastructure layer: autonomous systems that reason, decide, and act on real workloads using vLLM, LangGraph, and MCP.</description>
    </item>
    <item>
      <title>How KV-Cache Paging Works in vLLM — and Why It Matters for Production</title>
      <link>https://zhanyl-tech.github.io/blog/2026-01-15-kv-cache-paging-vllm/</link>
      <pubDate>Thu, 15 Jan 2026 00:00:00 +0000</pubDate>
      <guid>https://zhanyl-tech.github.io/blog/2026-01-15-kv-cache-paging-vllm/</guid>
      <description>A technical walkthrough of PagedAttention: the memory management innovation that makes vLLM practical for serving many concurrent LLM sessions without fragmenting GPU memory.</description>
    </item>
  </channel>
</rss>
