<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Context Engineering on Mi&amp;Bee Blog</title><link>/en/tags/context-engineering/</link><description>Recent content in Context Engineering on Mi&amp;Bee Blog</description><generator>Hugo -- gohugo.io</generator><language>en</language><managingEditor>蓝宝石的傻话</managingEditor><lastBuildDate>Sun, 10 Nov 2024 00:00:00 +0000</lastBuildDate><atom:link href="/en/tags/context-engineering/rss.xml" rel="self" type="application/rss+xml"/><item><title>From Prompts to Context: Why Clear Instructions Alone Are Not Enough</title><link>/en/posts/aihelper/from-prompt-to-context/</link><pubDate>Thu, 15 Aug 2024 00:00:00 +0000</pubDate><guid>/en/posts/aihelper/from-prompt-to-context/</guid><description>&lt;h2 id="a-frustrating-scenario"&gt;A Frustrating Scenario&lt;/h2&gt;
&lt;p&gt;Imagine this situation: you&amp;rsquo;re writing a perfect prompt about &amp;ldquo;the latest best practices for Python MySQL connections.&amp;rdquo; You carefully design a persona (&amp;ldquo;You are a Python database expert with 10 years of experience&amp;rdquo;), clear instructions (&amp;ldquo;Only provide 2024 best practices, no deprecated methods&amp;rdquo;), and specific format requirements (&amp;ldquo;List main approaches, pros/cons, code examples, security considerations&amp;rdquo;).&lt;/p&gt;
&lt;p&gt;This prompt is flawless. Yet, GPT confidently gives you code using 2018-deprecated methods with security vulnerabilities. Why?&lt;/p&gt;</description></item><item><title>Context Engineering: Giving AI the Right Knowledge</title><link>/en/posts/aihelper/context-engineering-guide/</link><pubDate>Sun, 10 Nov 2024 00:00:00 +0000</pubDate><guid>/en/posts/aihelper/context-engineering-guide/</guid><description>&lt;h2 id="what-is-context-engineering"&gt;What is Context Engineering?&lt;/h2&gt;
&lt;p&gt;In June 2025, Andrej Karpathy provided a definition of Context Engineering on the OpenAI engineering blog: &lt;strong&gt;&amp;ldquo;the delicate art and science of filling the context window with just the right information for the model to take the next step.&amp;rdquo;&lt;/strong&gt;&lt;/p&gt;
&lt;p&gt;This definition is exceptionally elegant. The core distinction from Prompt Engineering lies in:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Prompt Engineering&lt;/strong&gt;: Optimizes &amp;ldquo;what you say&amp;rdquo; – focuses on how input instructions are expressed&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Context Engineering&lt;/strong&gt;: Optimizes &amp;ldquo;what the model knows&amp;rdquo; – focuses on what information the model can access&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;Imagine this:&lt;/p&gt;</description></item></channel></rss>