<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Prompt-Engineering on ROBOCO</title>
    <link>https://roboco.io/en/tags/prompt-engineering/</link>
    <description>Recent content in Prompt-Engineering on ROBOCO</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Thu, 27 Nov 2025 07:57:49 +0900</lastBuildDate>
    <atom:link href="https://roboco.io/en/tags/prompt-engineering/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>LLM Prompting: Persona Assignment vs. Metacognition</title>
      <link>https://roboco.io/en/posts/persona-vs-metacognition/</link>
      <pubDate>Thu, 27 Nov 2025 07:57:49 +0900</pubDate>
      <guid>https://roboco.io/en/posts/persona-vs-metacognition/</guid>
      <description>&lt;blockquote&gt;&#xA;&lt;p&gt;On walking the tightrope between confidence and caution&lt;/p&gt;&#xA;&lt;/blockquote&gt;&#xA;&lt;figure&gt;&lt;img src=&#34;https://roboco.io/posts/images/Dohyun.png&#34;&gt;&lt;figcaption&gt;&#xA;      &lt;h4&gt;Dohyun Jung - Principal Consultant, ROBOCO&lt;/h4&gt;&#xA;    &lt;/figcaption&gt;&#xA;&lt;/figure&gt;&#xA;&#xA;&lt;h2 id=&#34;introduction&#34;&gt;Introduction&lt;/h2&gt;&#xA;&lt;p&gt;What is the difference between telling a model &amp;ldquo;answer like a senior engineer&amp;rdquo; and telling it &amp;ldquo;first think about what you are, then answer&amp;rdquo;? On the surface they look similar, but internally two completely different mechanisms are at work. This post examines the mathematical difference between the two approaches and what it means in practice.&lt;/p&gt;</description>
    </item>
    <item>
      <title>What Google&#39;s Prompt Engineering Whitepaper Says About Vibe Coding</title>
      <link>https://roboco.io/en/posts/google-prompt-engineering-whitepaper/</link>
      <pubDate>Sun, 13 Apr 2025 08:32:57 +0900</pubDate>
      <guid>https://roboco.io/en/posts/google-prompt-engineering-whitepaper/</guid>
      <description>&lt;figure&gt;&lt;img src=&#34;https://roboco.io/posts/images/Dohyun.png&#34;&gt;&lt;figcaption&gt;&#xA;      &lt;h4&gt;Dohyun Jung - Principal Consultant, ROBOCO&lt;/h4&gt;&#xA;    &lt;/figcaption&gt;&#xA;&lt;/figure&gt;&#xA;&#xA;&lt;h2 id=&#34;tldr&#34;&gt;TL;DR&lt;/h2&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;The code prompting section of Google&amp;rsquo;s prompt engineering whitepaper is a great starting point for using vibe coding at a professional level.&lt;/li&gt;&#xA;&lt;li&gt;Prompting for writing, explaining, translating, and debugging/reviewing code maps directly onto the vibe coding structure, where humans define the problem and AI carries it out.&lt;/li&gt;&#xA;&lt;li&gt;To get good results, spell out your output requirements, give positive instructions, and keep experimenting with and documenting reusable prompts.&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&lt;h2 id=&#34;introduction---prompt-engineering-and-vibe-coding&#34;&gt;Introduction - Prompt Engineering and Vibe Coding&lt;/h2&gt;&#xA;&lt;p&gt;Whenever I pick up a new technology or need to get up to speed on a topic, the first thing I read is the whitepaper published by the vendor behind it. Most technical whitepapers are free, and they pack in a lot of information, which saves me time.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
