<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>MCP on Umut Doğan</title><link>https://umutdogan.com/tags/mcp/</link><description>Recent content in MCP on Umut Doğan</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sat, 05 Sep 2026 09:00:00 +0300</lastBuildDate><atom:link href="https://umutdogan.com/tags/mcp/index.xml" rel="self" type="application/rss+xml"/><item><title>Model Context Protocol (MCP) Explained: Theory, Architecture, and a Java + Spring Boot + Docker Tutorial</title><link>https://umutdogan.com/posts/2026-09-05-model-context-protocol-java-tutorial/</link><pubDate>Sat, 05 Sep 2026 09:00:00 +0300</pubDate><guid>https://umutdogan.com/posts/2026-09-05-model-context-protocol-java-tutorial/</guid><description>&lt;p&gt;I&amp;rsquo;ve written before about &lt;a href="https://umutdogan.com/posts/2024-12-20-agentic-ai/" &gt;agentic AI&lt;/a&gt; and &lt;a href="https://umutdogan.com/posts/2025-03-15-spring-ai-for-beginners/" &gt;building with Spring AI&lt;/a&gt;. Both of those posts touched on a problem that keeps coming up once you move past toy demos: how does an AI application actually get access to &lt;em&gt;your&lt;/em&gt; data and &lt;em&gt;your&lt;/em&gt; tools, in a way that isn&amp;rsquo;t a bespoke integration for every single one? That&amp;rsquo;s the problem the &lt;strong&gt;Model Context Protocol (MCP)&lt;/strong&gt; sets out to solve.&lt;/p&gt;&#10;&lt;p&gt;This post is in two parts. First, the theory — what MCP actually is, its architecture, and the concepts you need before anything else makes sense. Then a full hands-on tutorial: we&amp;rsquo;ll build a real MCP server in Java with Spring Boot, run it, call it with raw JSON-RPC over curl, and package it up with Docker.&lt;/p&gt;</description></item></channel></rss>