<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Sufyan Ahmad — Blog</title>
    <link>https://www.sufyanahmad.com</link>
    <description>Articles on AI, machine learning, NLP, and engineering from Sufyan Ahmad.</description>
    <language>en</language>
    <lastBuildDate>Fri, 08 May 2026 21:48:21 GMT</lastBuildDate>
    <atom:link href="https://www.sufyanahmad.com/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Building Production Multi-Agent Systems with LangGraph</title>
      <link>https://www.sufyanahmad.com/blog/building-production-multi-agent-systems-langgraph</link>
      <guid isPermaLink="true">https://www.sufyanahmad.com/blog/building-production-multi-agent-systems-langgraph</guid>
      <pubDate>Thu, 20 Mar 2025 00:00:00 GMT</pubDate>
      <description>How I moved beyond single-chain LLM apps to stateful, fault-tolerant multi-agent pipelines—the architectural decisions, failure modes, and lessons learned shipping LangGraph systems to real users.</description>
    </item>
    <item>
      <title>Voice AI in Production: Real-Time Agents with LiveKit and ElevenLabs</title>
      <link>https://www.sufyanahmad.com/blog/voice-ai-production-livekit-elevenlabs</link>
      <guid isPermaLink="true">https://www.sufyanahmad.com/blog/voice-ai-production-livekit-elevenlabs</guid>
      <pubDate>Fri, 14 Feb 2025 00:00:00 GMT</pubDate>
      <description>Building voice agents that actually work in production means solving latency, interruption handling, and turn-taking at the same time. Here&apos;s what I learned building real-time conversational AI with LiveKit, streaming STT, and ElevenLabs TTS.</description>
    </item>
    <item>
      <title>RAG That Actually Works: From Chunking Strategy to Production Retrieval</title>
      <link>https://www.sufyanahmad.com/blog/rag-production-chunking-retrieval-qdrant</link>
      <guid isPermaLink="true">https://www.sufyanahmad.com/blog/rag-production-chunking-retrieval-qdrant</guid>
      <pubDate>Wed, 08 Jan 2025 00:00:00 GMT</pubDate>
      <description>Most RAG tutorials show you the happy path—clean documents, obvious queries, perfect retrieval. Production is nothing like that. Here&apos;s what I&apos;ve learned building RAG systems over 50K documents that serve real users.</description>
    </item>
  </channel>
</rss>