<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Week 9 - Agent Memory, Tools &amp; MCP on AI Platform Engineering Handbook</title><link>/docs/week-09/</link><description>Recent content in Week 9 - Agent Memory, Tools &amp; MCP on AI Platform Engineering Handbook</description><generator>Hugo</generator><language>en</language><copyright>Copyright (c) 2026 Harshhaa</copyright><atom:link href="/docs/week-09/index.xml" rel="self" type="application/rss+xml"/><item><title>Agent Memory Systems</title><link>/docs/week-09/agent-memory-systems/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>/docs/week-09/agent-memory-systems/</guid><description>&lt;hr&gt;
&lt;h2 id="1-short-term-memory"&gt;1. Short-term Memory&lt;/h2&gt;
&lt;p&gt;Short-term memory refers to the information an agent needs to keep track of just within its current, ongoing task or conversation — everything that&amp;rsquo;s happened so far in this particular session, held onto just long enough to be useful for the immediate work at hand, without necessarily needing to be remembered forever afterward.&lt;/p&gt;</description></item><item><title>Model Context Protocol (MCP)</title><link>/docs/week-09/model-context-protocol-mcp/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>/docs/week-09/model-context-protocol-mcp/</guid><description>&lt;hr&gt;
&lt;h2 id="1-mcp-fundamentals"&gt;1. MCP Fundamentals&lt;/h2&gt;
&lt;p&gt;At its core, MCP is an open protocol — meaning a publicly available, standardized set of rules — that defines a consistent way for AI applications to receive context and capabilities from external systems. The word &amp;ldquo;protocol&amp;rdquo; is doing important work here: much like how HTTP is the standardized protocol that lets any web browser talk to any website regardless of who built either one, MCP aims to be a standardized way for any AI application to talk to any external tool or data source, regardless of who built either side.&lt;/p&gt;</description></item><item><title>Tool Calling</title><link>/docs/week-09/tool-calling/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>/docs/week-09/tool-calling/</guid><description>&lt;hr&gt;
&lt;h2 id="1-function-calling"&gt;1. Function Calling&lt;/h2&gt;
&lt;p&gt;Function calling is the foundational mechanism that makes tool use possible in the first place, so it&amp;rsquo;s worth understanding clearly before anything else in this section. At its heart, the idea is this: instead of a language model only being able to respond with plain, free-form text, it can also be given the ability to respond by saying, in a very specific, structured way, &amp;ldquo;I want to call this particular function, with these particular input values.&amp;rdquo;&lt;/p&gt;</description></item></channel></rss>