<?xml version="1.0"?>
<rss version="2.0">
   <channel>
      <title>Prompt engineering by Elizabeth Georgiades</title>
      <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j</link>
      <description>activities </description>
      <language>en-us</language>
      <pubDate>2025-09-09 08:44:02 UTC</pubDate>
      <lastBuildDate>2025-09-09 10:30:08 UTC</lastBuildDate>
      <webMaster>hello@padlet.com</webMaster>
      <image>
         <url>https://padlet.net/icons/png/1f4c8.png</url>
      </image>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575678400</link>
         <description><![CDATA[<p>Prompt for you to use:</p><p><br></p><p>Here is an AI-generated output. Your task is to reconstruct the original prompt that could have produced this. Pay attention to style, tone, structure, and detail. Write your version of the prompt, test it with the AI, and compare the outputs.</p><p><br></p><p>Sample Output to Start With (for practice):</p><p>"<mark>In the future, cities will be powered by renewable energy, with autonomous vehicles reducing traffic and pollution. Communities will be connected through advanced digital networks that make information, healthcare, and education accessible to all</mark>."</p>]]></description>
         <enclosure url="" />
         <pubDate>2025-09-09 08:45:07 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575678400</guid>
      </item>
      <item>
         <title>You’ll be given an AI-generated output (it could be text, an image, or a short piece of code). Your challenge is to figure out what the original prompt might have been.</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575679343</link>
         <description><![CDATA[<p>What to do:</p><p><br/></p><ul><li><p>Look carefully at the style, tone, level of detail, and structure of the output.</p></li><li><p>Write a prompt that you think could have produced it.</p></li><li><p>Test your prompt with the AI and compare your output to the original.</p></li></ul><p><br/></p><p><mark>Why this matters:</mark></p><p>This activity helps you understand how the way you phrase a prompt shapes the AI’s response. It also trains you to “think backwards” about AI behaviour.</p>]]></description>
         <enclosure url="" />
         <pubDate>2025-09-09 08:45:45 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575679343</guid>
      </item>
      <item>
         <title>You’ll see an AI output that includes stereotypes, missing perspectives, or shallow reasoning. Your job is to spot the problem and fix it.</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575680281</link>
         <description><![CDATA[<p>What to do:</p><p><br/></p><ul><li><p>Identify what’s biased, limited, or unbalanced in the output.</p></li><li><p>Rewrite the original prompt to reduce the problem (e.g., ask for diverse perspectives, avoid gendered language, or request evidence).</p></li><li><p>Run your new prompt and compare the outputs.</p></li></ul><p><br/></p><p><mark>Why this matters:</mark></p><p>This helps you build awareness of bias in AI and practice how to write prompts that create fairer, more accurate, and more inclusive results.</p>]]></description>
         <enclosure url="" />
         <pubDate>2025-09-09 08:46:20 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575680281</guid>
      </item>
      <item>
         <title>You’ll work with a problem that needs reasoning (a logic puzzle, a coding bug, or a tricky decision). Instead of asking the AI for the answer in one go, your challenge is to design a prompt that makes it show its thinking step by step.</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575681503</link>
         <description><![CDATA[<p>What to do:</p><p><br/></p><ul><li><p>Break the problem into smaller steps.</p></li><li><p>Write a prompt that asks the AI to solve the problem one step at a time (“First explain the concept, then solve, then check your answer…”).</p></li><li><p>Test and refine your prompt until the reasoning is clear and accurate.</p></li></ul><p><br/></p><p><mark>Why this matters:</mark></p><p>This activity trains you to think like a problem-solver and to guide AI reasoning in ways that are transparent and reliable.</p>]]></description>
         <enclosure url="" />
         <pubDate>2025-09-09 08:46:57 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575681503</guid>
      </item>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575682165</link>
         <description><![CDATA[<p>Prompt for you to use:</p><p><br/></p><p>Here is an AI output that may include bias or missing perspectives. Your task is to identify the issue, then rewrite the prompt to reduce bias and generate a more balanced response. Run the new prompt and compare.</p><p><br/></p><p>Sample Output to Start With (for practice):</p><p>"<mark>A good leader is always strong, confident, and decisive. He must be able to take charge and guide his team to success."</mark></p>]]></description>
         <enclosure url="" />
         <pubDate>2025-09-09 08:47:14 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575682165</guid>
      </item>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575682989</link>
         <description><![CDATA[<p>Prompt for you to use:</p><p><br/></p><p>Your task is to design a prompt that forces the AI to solve a problem step by step. Break the problem into smaller parts, and tell the AI to explain its reasoning before giving a final answer. Test and refine until the reasoning is clear.</p><p><br/></p><p>Sample Challenge to Start With:</p><ul><li><p>"<mark>What is 17 × 23?"</mark></p></li><li><p><mark>First prompt: “Solve 17 × 23 step by step. Show your reasoning before giving the final answer.”</mark></p></li><li><p><mark>Then refine: “First break the problem into smaller multiplications, then add the results together, then check your calculation before giving the final answer.</mark>”</p></li></ul>]]></description>
         <enclosure url="" />
         <pubDate>2025-09-09 08:47:47 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575682989</guid>
      </item>
      <item>
         <title>Mario</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575774798</link>
         <description><![CDATA[<p>Strong answer Mario - your output matches tone and structure of the original example we used and shows you picked up on the futuristic optimism and the urban planning focus. </p>]]></description>
         <enclosure url="https://media3.giphy.com/media/v1.Y2lkPWNhYmM5OTE4MjkxOHR0eGx1dHlzc2o5NjF0MmVsMWtlcTY1M3Z3Z3BkazUwbHRxeCZlcD12MV9naWZzX3NlYXJjaCZjdD1n/yyZRSvISN1vvW/giphy.gif" />
         <pubDate>2025-09-09 09:52:17 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575774798</guid>
      </item>
      <item>
         <title>Gialdino</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575776889</link>
         <description><![CDATA[<p>Gialdino, your output shows strong critical thinking. You shifted the scope and demonstrated how specific prompts shape thematic AI outputs. You used word-limits (nice!) that mirror real world prompt design. Your output is a little longer than Mario's and shows an added layer of ethical and social dimension. </p>]]></description>
         <enclosure url="https://media2.giphy.com/media/v1.Y2lkPWNhYmM5OTE4bmRiamViejVwdDgyZTlscmo0ejZoNml0cHk5ZWZtMGVzY2xjeWJidyZlcD12MV9naWZzX3NlYXJjaCZjdD1n/ZdO4NenDbsQNwUwKDB/giphy.gif" />
         <pubDate>2025-09-09 09:54:12 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575776889</guid>
      </item>
      <item>
         <title>Sebastian</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575783935</link>
         <description><![CDATA[<p>You've done a great job in setting up the prompt by linking to everyday problems, this shows you have a strong awareness of how to give AI context. Your prompt has breadth and coherence; it flows and is logical, demonstrating effective prompt design sequence. </p><p><br></p><p>Area to improve: slight tweaking to this part of your prompt to give AI greater clarity 'how future cities will function and how communities will be connected'. Give your AI prompt a constraint like a word count - write a short prediction is a bit vague for AI and may give you more information than you asked for. Consider seperating the tasks from your prompt, like First cities, then communities this might help you control the output better. </p><p><br></p><p><br></p>]]></description>
         <enclosure url="https://media1.giphy.com/media/v1.Y2lkPWNhYmM5OTE4eDB6MG12Z2FqaGV2dzgwYXVpOWxqOWlxeHJmMWVseGdpY2xlOXB4dyZlcD12MV9naWZzX3NlYXJjaCZjdD1n/vZJAGBERy0Lp7NFyA2/giphy.gif" />
         <pubDate>2025-09-09 10:00:07 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575783935</guid>
      </item>
      <item>
         <title>Pamela</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575787195</link>
         <description><![CDATA[<p>Really clear framing Pamela, your prompt is explicit about the conditions (renewable energy etc) and includes a constraint like a word limit. Consider splitting your input into two prompts to see if you get richer details. Sometimes we might overload the input and create scope creep on how much information we want. </p>]]></description>
         <enclosure url="https://media0.giphy.com/media/v1.Y2lkPWNhYmM5OTE4MjJ4bnM5a2VheThoZWtveDM0a3VibWpnMGFta3hqem1iOGM4ajFmZiZlcD12MV9naWZzX3NlYXJjaCZjdD1n/wQVQOSYCUM1zXOm1AB/giphy.gif" />
         <pubDate>2025-09-09 10:03:12 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575787195</guid>
      </item>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575790913</link>
         <description><![CDATA[<p>Great analysis of the bias in the prompt. This kind of awareness is what you should be aiming for when working with AI. You've both used AI to reprompt and reduce the bias, really great job. </p>]]></description>
         <enclosure url="https://media2.giphy.com/media/v1.Y2lkPWNhYmM5OTE4eGkyMWVtb3FkdXE4YnZlbzZpMHl1MXpsdzV2ZnI4c21ncDM4cnJhNSZlcD12MV9naWZzX3NlYXJjaCZjdD1n/mD9DcTbJq4QYPV4gMn/giphy.gif" />
         <pubDate>2025-09-09 10:06:48 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575790913</guid>
      </item>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575792237</link>
         <description><![CDATA[]]></description>
         <enclosure url="https://mymagicprompt.com/" />
         <pubDate>2025-09-09 10:07:54 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575792237</guid>
      </item>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575793057</link>
         <description><![CDATA[]]></description>
         <enclosure url="https://promptperfect.jina.ai/" />
         <pubDate>2025-09-09 10:08:29 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575793057</guid>
      </item>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575793630</link>
         <description><![CDATA[]]></description>
         <enclosure url="https://flowgpt.com/" />
         <pubDate>2025-09-09 10:08:57 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575793630</guid>
      </item>
      <item>
         <title></title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575793985</link>
         <description><![CDATA[]]></description>
         <enclosure url="https://prompthero.com/" />
         <pubDate>2025-09-09 10:09:22 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575793985</guid>
      </item>
      <item>
         <title>Tania</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575797278</link>
         <description><![CDATA[<p>Tania, you’ve done a great job with this response. I like how you turned the AI’s output into clear class topics and activities. The structure makes it easy to see how each idea (energy, vehicles, traffic, pollution, healthcare, education) could be taught in class. Highlighting the “best topics” for engagement was also a smart move.</p><p>To make it even stronger, try tightening your prompt wording (some phrases repeat) and think about adding limits, like asking the AI for “one debate and one experiment per topic.” This helps control the output and makes it easier to compare.</p><p>Overall, this shows you can use AI not just to answer but to build useful teaching ideas. I might steal your idea! Well done!</p>]]></description>
         <enclosure url="https://media2.giphy.com/media/v1.Y2lkPWNhYmM5OTE4d3g4anA1OGdtcDhidzkyOTk4Y3I1Njdsd28xNGlxazBkcHluNTJsdiZlcD12MV9naWZzX3NlYXJjaCZjdD1n/f643FJfVotwsuDERhM/giphy.gif" />
         <pubDate>2025-09-09 10:12:45 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575797278</guid>
      </item>
      <item>
         <title>Tania</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575800077</link>
         <description><![CDATA[<p>You’ve done a great job breaking the multiplication into clear steps and then asking for a “speech” to explain it — that shows strong awareness of how AI can support learning, not just give answers. To refine this further, try cleaning up the grammar in your prompt (e.g., “cautelous” → “cautious”), as clearer wording usually leads to sharper AI outputs. You could also add limits (like “use numbered steps” or “keep under 80 words”) to make the response easier for students to follow.</p>]]></description>
         <enclosure url="https://media1.giphy.com/media/v1.Y2lkPWNhYmM5OTE4ZHVieG1qYTdsZW11emp5cTR4bWFvMHM3Y3I2ZGZlNTl1cXRlcjVyMiZlcD12MV9naWZzX3NlYXJjaCZjdD1n/g9582DNuQppxC/giphy.gif" />
         <pubDate>2025-09-09 10:15:39 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575800077</guid>
      </item>
      <item>
         <title>Mario</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575801866</link>
         <description><![CDATA[<p>Mario, this is an excellent breakdown. You not only split 17 and 23 into smaller parts but also showed the distributive property clearly, which makes the logic very transparent for learners. Each step is correct and well-sequenced. To refine it even further, you could simplify the formatting  for example, some numbers are repeated (e.g., “391391”) which might distract the reader. Consider, adding a brief, student-friendly explanation (“why breaking into parts makes it easier”) would make it even stronger as a teaching example.</p>]]></description>
         <enclosure url="https://media4.giphy.com/media/v1.Y2lkPWNhYmM5OTE4M2toMHF3cG1uYWlhMnJweGZtdTQ1MnNpcjRjZWZseXV3bzh3aWY3cCZlcD12MV9naWZzX3NlYXJjaCZjdD1n/1236TCtX5dsGEo/giphy.gif" />
         <pubDate>2025-09-09 10:17:16 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575801866</guid>
      </item>
      <item>
         <title>Bastien</title>
         <author>lxdintheloopdesign</author>
         <link>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575803657</link>
         <description><![CDATA[<p>Bastien, your solution is very clear and easy to follow. Breaking 23 into 20 + 3 was a smart choice, and you showed the smaller multiplications step by step before adding them back together. I like that you also included a “check” step because this reinforces good problem-solving habits. To make it even stronger, you could explain <em>why</em> breaking the number down makes the multiplication easier for learners, not just how to do it. That extra teaching touch would turn this from a worked solution into a learning experience.</p>]]></description>
         <enclosure url="https://media0.giphy.com/media/v1.Y2lkPWNhYmM5OTE4dXVwM2l6cWdodnltMTE5MXZpb3QwbDljYXQxejg3cmF4ZGxnMjVkbSZlcD12MV9naWZzX3NlYXJjaCZjdD1n/l3q2XhfQ8oCkm1Ts4/giphy.gif" />
         <pubDate>2025-09-09 10:18:50 UTC</pubDate>
         <guid>https://padlet.com/lxdintheloopdesign/tsi677b1rjkkxp3j/wish/3575803657</guid>
      </item>
   </channel>
</rss>
