<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: chundrusumanth</title>
    <description>The latest articles on DEV Community by chundrusumanth (@chundrusumanth).</description>
    <link>https://dev.to/chundrusumanth</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4025827%2Fd48e3cdd-3946-4bee-892c-60a214fe9a76.jpg</url>
      <title>DEV Community: chundrusumanth</title>
      <link>https://dev.to/chundrusumanth</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/chundrusumanth"/>
    <language>en</language>
    <item>
      <title>How I built a Multi Agent Course Curation System with A2A Protocol and GCP</title>
      <dc:creator>chundrusumanth</dc:creator>
      <pubDate>Mon, 13 Jul 2026 06:40:57 +0000</pubDate>
      <link>https://dev.to/chundrusumanth/how-i-built-a-multi-agent-course-curation-system-with-a2a-protocol-and-gcp-5b0e</link>
      <guid>https://dev.to/chundrusumanth/how-i-built-a-multi-agent-course-curation-system-with-a2a-protocol-and-gcp-5b0e</guid>
      <description>&lt;p&gt;*This project demonstrates a multi-agent system built with the Google Cloud Agent Development Kit (ADK) that automates the creation of educational content. The system uses a Researcher agent, a strict Judge agent for quality control, and a Content Builder agent for formatting, resulting in a robust, modular workflow. The key advantage of this architecture lies in the separation of concerns, which enhances debugging and ensures reliable output through structured validation. This post is my submission for&lt;br&gt;
This project solves the issue of the online course curation and search using multiple prompts with a single AI Agent&lt;br&gt;
I have created a separate agent such as a research agent for researching and finding the up to date information about  particular course and a judeg agent which can judge the research agents work and gives feedback and based on the research agent will decide whether run the feedback loop or not based on the judge agents response and finally I have also created the orchestrator agent which manage the entire workflow of the agents and give the information for the user with the accurate up to date information for which the agentic system is created for.&lt;/p&gt;

&lt;h2&gt;
  
  
  Key Learnings
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;This challenge taught me how to use A2A protocol with multiple agents.&lt;/li&gt;
&lt;li&gt;This is my first project on GCP and it also taught me how to use GCP console and resources for creating and deploying the application using google cloud.&lt;/li&gt;
&lt;li&gt;This also taught me how to create a multi agent systems using GCP and A2A protocol.&lt;/li&gt;
&lt;/ol&gt;

</description>
      <category>agents</category>
      <category>buildmultiagents</category>
      <category>gemini</category>
      <category>adk</category>
    </item>
  </channel>
</rss>
