<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Dag on Nitin Kumar Singh</title><link>https://nitinksingh.com/tags/dag/</link><description>Recent content in Dag on Nitin Kumar Singh</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>© 2026 Nitin Kumar Singh. All rights reserved.</copyright><lastBuildDate>Sun, 12 Apr 2026 11:05:00 +0530</lastBuildDate><atom:link href="https://nitinksingh.com/tags/dag/index.xml" rel="self" type="application/rss+xml"/><item><title>Graph-Based Workflows -- Beyond Simple Orchestration</title><link>https://nitinksingh.com/posts/graph-based-workflows--beyond-simple-orchestration/</link><pubDate>Sun, 12 Apr 2026 11:05:00 +0530</pubDate><guid>https://nitinksingh.com/posts/graph-based-workflows--beyond-simple-orchestration/</guid><description>&lt;p&gt;Throughout this series, we have relied on a single orchestration pattern: the LLM decides what to do next. The orchestrator receives a user message, its system prompt teaches it which specialists exist, and the model picks the right one. For most interactions &amp;ndash; &amp;ldquo;search for headphones,&amp;rdquo; &amp;ldquo;what is my order status,&amp;rdquo; &amp;ldquo;any coupons available?&amp;rdquo; &amp;ndash; this works well. The LLM routes accurately, the specialist responds, and the orchestrator formats the answer.&lt;/p&gt;</description></item></channel></rss>