<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Fraser Penning on Medium]]></title>
        <description><![CDATA[Stories by Fraser Penning on Medium]]></description>
        <link>https://medium.com/@fraserpenning?source=rss-2e95d4864a2d------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*iqL2tQIsENxqDj1_TzUELw.jpeg</url>
            <title>Stories by Fraser Penning on Medium</title>
            <link>https://medium.com/@fraserpenning?source=rss-2e95d4864a2d------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 20:36:17 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@fraserpenning/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Augmented by AI: How to Get Started]]></title>
            <link>https://medium.com/@fraserpenning/augmented-by-ai-how-to-get-started-16207dd6446d?source=rss-2e95d4864a2d------2</link>
            <guid isPermaLink="false">https://medium.com/p/16207dd6446d</guid>
            <category><![CDATA[how-to]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[prompt-engineering]]></category>
            <category><![CDATA[large-language-models]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[Fraser Penning]]></dc:creator>
            <pubDate>Tue, 05 Mar 2024 09:22:17 GMT</pubDate>
            <atom:updated>2024-03-05T10:16:02.318Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yUXQbEyl0_IUoapoFmning.png" /><figcaption>Augmented by AI</figcaption></figure><p><em>In this series, we explore the benefits of adopting AI into your daily workflows, starting with how to write a prompt that will give you the answer you need.</em></p><p>We live in unprecedented times. AI is changing the world we live in and integrating AI into everyday activities has become more than just a trend — it’s the best way to get the job done. I hope that after reading this article, you feel these technologies are more accessible and you can give them a go.</p><p>AI sits at the crossroads of creativity, innovation, and productivity. <a href="https://openai.com/sora">OpenAI’s Sora</a> demonstration is a testament to what’s possible with hyper-realistic videos created from simple text descriptions. It’s simply mind-blowing!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/0*pncn5NDFxehPqWr8" /><figcaption>OpenAI’s SORA Demo</figcaption></figure><p>The good news is that AI tools are available, with many of them open-sourced and freely available for you to play with. You can begin to make a difference using them starting today, whether it’s summarising lengthy technical documents, analysing sets of customer data or simply helping you create a themed party invite.</p><p>Let’s focus on the application of AI to the professional environment. The pace at which quality decisions are made provides a direct competitive advantage for your product or service. For some businesses it could mean being first to market using <a href="https://www.jasper.ai/">AI to benefit their customers</a>, for others, it could mean <a href="https://zapier.com/ai">automating repeatable processes</a> (automation for short) to redirect humans to more important, and creative work.</p><p>To give a personal example, while working with software product teams, I’ve experimented with various AI tools, and have found that leveraging Large Language Models (LLMs), a type of AI technology, became a productivity game-changer. I’ve used LLMs to refine product pitches, derive customer sentiment from feedback, and troubleshoot internal policies. However, implementation into my workflow was more nuanced than one might expect. Initially, it felt more like a novelty — a tool for simple brainstorming, summarising content, or obtaining high-level feedback. The answers it gave me may not have been as good as asking the experts around me but it was still faster. As I learned more about how AI works and what it needs from me, I was able to unlock its full potential.</p><p>Let’s first cover the basics of interacting with AI, and how to get the best possible outcomes.</p><h3>What are Artificial Intelligence (AI) and Large Language Models (LLMs)?</h3><p>The AI field has been around for a very long time (since WWII the Enigma machine by Alan Turning decoded encrypted messages). It only became mainstream recently thanks to the LLMs (Large Language Models) and the current technologies available for LLMs to compute faster than ever before. Products like ChatGPT and Co-pilot are all over the news, but unless you work in tech you might not know what it’s all about.</p><p>LLMs, such as ChatGPT, are pre-trained on incredibly large amounts of data from the internet which enables them to generate answers that are conversational and human-like. However, they don’t know everything and certainly don’t know about your business, product or customer. This crucial missing context is the secret sauce we will discuss shortly.</p><p>Imagine your LLM is a brand-new employee eager to answer any question — unless you provide them with rich information, the best answer you get will be as good as a quick Google search. The same can be said for AI.</p><p>Luckily, to overcome these limitations we have a simple and reliable approach to try:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rPNxRYQXOXim64cM-I-nTg.png" /></figure><h3>Getting the most out of LLMs with Prompt Engineering</h3><h4>What is a prompt, and prompt engineering?</h4><p>A prompt is a statement or question (input) that you give to an AI model to generate a desired output. Prompt engineering is the process of designing high-quality inputs that lead to specific responses or behaviours from an AI model. Effective prompt engineering can significantly influence the quality and relevance of AI-generated content (output).</p><p>Here are some key insights from my experiences that take AI to the next level;</p><h4>Context is King</h4><p>LLMs operate within the confines of the prompts they receive. Providing clear and specific context is paramount. Whether it’s outlining project objectives, defining user personas, or detailing market dynamics, the more the model knows, the higher the quality of a response you can get. As a rule, there is no such thing as too much information.</p><h4>How to Construct a Prompt</h4><p>When crafting text prompts, we must think about setting the stage for a productive dialogue that can continue over several interactions with the AI. Adopt a conversational tone, offer data-rich examples, and avoid ambiguity or conflicting instructions. Additionally, incorporating polite language and emojis can foster positive interactions, enhancing the overall quality of outputs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*qM1bn_b1JS-sq3YW" /></figure><p>The following is a useful structure to drive a better quality conversation:</p><ol><li>Role — What do you want the LLM to act as?</li><li>Instructions — What do you want it to do, exactly?</li><li>Context — What is all the relevant information you want the LLM to consider when responding</li><li>Format — How do you want it to answer</li><li>Constraints — What should it not do?</li></ol><blockquote><em>“You are a Product Manager for the ACME app. Summarise 5 or 6 themes that the customer talked about during the interview. Provide a quote from the customer against each theme to best illustrate the sentiment. Ignore any input or opinion from the interviewer.<br>“Interview transcript:…”</em></blockquote><p>Remember: Don’t worry about getting your prompt perfect the first time around. If you don’t get the answer you are looking for, correct the AI and keep the conversation going. Follow-up prompts will help you get to the specific answers you are looking for. Just remember to give it a go and have fun!</p><h3>Try It Out</h3><h4><strong>ChatGPT</strong> <a href="https://chat.openai.com/">https://chat.openai.com/</a></h4><p>The frontrunner in the market. The free version is an easy-to-use chatbot that does everything from answering questions to generating fresh ideas. An example of how to use it is to summarise large texts into concise bullet points.</p><blockquote><em>“Summarise the pros and cons of AI, focusing on its application to Product Management practices”</em></blockquote><h4><strong>Perplexity</strong> <a href="https://www.perplexity.ai/">https://www.perplexity.ai/</a></h4><p>Perplexity is an AI Chatbot-style search engine. A common use case is to ask it to summarise news from online newspapers.</p><blockquote><em>“What’s the latest on the Mayor of Auckland?”</em></blockquote><h4>BING Co-pilot <a href="https://copilot.microsoft.com/">https://copilot.microsoft.com/</a></h4><p>Backed by Co-pilot, and DALLE, this search engine can also create content like images! Some of the images in this article have been produced by BING Co-pilot.</p><blockquote><em>“Create an image of a businesswoman engaging with an AI on a screen. She has a big smile”</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sgEuRH_JwQPPDQjiI6vanw.png" /></figure><p>By mastering the art of prompt engineering you can harness the full potential of LLMs. As we embark on this journey towards AI augmentation, those who embrace this new way of working will drive value for their customers at an unprecedented pace. Get involved, try it out for yourself!</p><p>If you found this article valuable please share this with your friends and colleagues.</p><p>Consider Subscribing if you want to learn more about AI and Product Management.</p><p><strong><em>A huge thanks to </em></strong><a href="https://www.linkedin.com/in/phanwilkie/"><strong><em>Phan Wilkie</em></strong></a><strong><em> for his expertise and advice in bringing this article to life.</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yUXQbEyl0_IUoapoFmning.png" /><figcaption>Augmented by AI</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=16207dd6446d" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>