<?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 Justin Hevey on Medium]]></title>
        <description><![CDATA[Stories by Justin Hevey on Medium]]></description>
        <link>https://medium.com/@spl72?source=rss-faf3a54a5d03------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*M57Kq5_w13Mz-jsk.png</url>
            <title>Stories by Justin Hevey on Medium</title>
            <link>https://medium.com/@spl72?source=rss-faf3a54a5d03------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 19:35:35 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@spl72/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[Designing for the Future: Navigating the Landscape of AI Agents and Human-Machine Interaction]]></title>
            <link>https://medium.com/@spl72/designing-for-the-future-navigating-the-landscape-of-ai-agents-and-human-machine-interaction-9cf2b0c2fd8a?source=rss-faf3a54a5d03------2</link>
            <guid isPermaLink="false">https://medium.com/p/9cf2b0c2fd8a</guid>
            <category><![CDATA[human-comuter-interaction]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[future]]></category>
            <category><![CDATA[design-principles]]></category>
            <dc:creator><![CDATA[Justin Hevey]]></dc:creator>
            <pubDate>Fri, 17 May 2024 09:31:40 GMT</pubDate>
            <atom:updated>2024-05-17T09:31:40.008Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jUjKar5hX5I_CJm6p0pYtg.png" /><figcaption>Midjourney</figcaption></figure><p>I’ve recently just finished reading <a href="https://en.wikipedia.org/wiki/The_Innovators_(book)">The Innovators by Walter Isaacson</a>. It is a complete timeline of how digital machines came about to what we have today. There is so much in this book to unpack on where we are today and what might be next. It’s an excellent read, and it’s hard to fathom how far we have come in the last 50 years alone.</p><p>In the book, Walter Isaacson touches on Douglas Engelbert’s “Mother of All Demos” where <a href="https://en.wikipedia.org/wiki/Douglas_Engelbart"><strong>Douglas Engelbart</strong></a> in 1968, describes exactly how humans would interact and gain value from using machines. Engelbert’s work ended up being developed into a 45k-word manifesto of his vision (<a href="https://web.archive.org/web/20110504035147/http://www.dougengelbart.org/pubs/augment-3906.html">Link</a>). Published 1962 — <em>Augmenting Human Intellect: A Conceptual Framework</em>. His premise here was not to replace human thought with machines but argued and believed that the intuitive talent of the human mind should be combined with the processing abilities of the machines to create an integrative domain where hunches, trial-and-error methods, and an intuitive human understanding of situations can coexist with machines.</p><p>Engelbert’s vision, although not directly related to developing AI algorithms or machine learning models, is remarkably relevant in our current AI landscape. His ideas have shaped the way humans interact with machines, and AI is a natural extension of this true human-machine interaction augmentation.</p><blockquote><strong>As computation (AI) changes these relationships, these interfaces and interactions will change, too. Our design needs to change with it — what we design, how we design, where we design, and who we design with.<br> — Emily Campbell — </strong><a href="https://shapeofai.substack.com/"><strong>Shape of ai/UX</strong></a></blockquote><p>Emily writes a great piece on ‘How do we define good AI design?’ What frameworks and decisions do we as designers need to consider? What principles should be considered? How does a systems thinking approach provide hints to the experience? Designers really can shape these conversations and write the rules here.</p><p>Many conversations focus on technology or improving human productivity, but we should dive deeper to discover and explore new use cases and workflows. As designers and future thinkers, we serve as a bridge connecting past experiences with future developments. We evaluate what strategies were effective, which failed, and how we’ve provided value in connecting the shift in paradigms. Now is the time to reinterpret these experiences in the context of human-computer interaction. As adaptable individuals, designers are well-positioned to navigate these new challenges.</p><h4>Emily states what good AI design needs:</h4><ul><li>It has to be usable.</li><li>It has to be familiar, fluid, and understandable.</li><li>It has to be ethical and trustworthy.</li></ul><p>There is a missing piece in her mind: The relationship to AI has changed.<em> </em>Emily provides solid thinking about how this relationship is changing and how we can measure it as we design for and with AI — heuristics for AI design. <a href="https://www.linkedin.com/pulse/my-emerging-heuristics-assessing-ai-design-emily-campbell-f2zqc/?trackingId=aNAuSNCHQr2MOo6yvht91Q%3D%3D">Read here full article</a></p><p>This leads me to what I have been exploring lately: AI agents. What are they? When will we start using them? What are the benefits and types, and how is design placed to help design these experiences?</p><h3>AI agents &amp; agentic systems</h3><p>Most people today use GenAI technology for zero-shot or few-shot prompting interactions. It takes a lot of trial and error to get what we want from our initial thoughts or expectations into intended outcomes. It’s a bit like using a computer without online access. We ask, hoping the machine’s model fully understands our imagination and is in sync with our thoughts.</p><p>AI agents will change how we interact with machines; this week, ChatGPT announcements delivered new multi-modal interactions, which were amazing. It will be like every person will have a Personal Assistant (PA). They will probably have many PAs, which will all interact with each other. Imagine you have the job of organising a large group of people in a given location. There are flights, travel, the need to find somewhere where we’re going to meet, activities for kids, things to do, etc. All this can be done with agents.</p><p><strong>Let’s look at the types of AI agents that we’ll soon see.</strong></p><ul><li><strong>Simple Reflex agents:</strong> These agents react solely to current inputs via simple rules and are ideal for stable, straightforward environments, like a rule-based automated customer support system.</li><li><strong>Model-based Reflex agents:</strong> Use an internal state that integrates percept history with an environmental model, enabling them to make inferences and effectively operate in partially observable environments.</li><li><strong>Goal-based agents:</strong> These agents aim for a specific goal, evaluating current conditions and potential actions to select the best strategic path to success.</li><li><strong>Utility-based agents:</strong> Operate like goal-based ones but decide actions based on a utility function, aiming to maximise expected utility instead of merely achieving a goal state.</li><li><strong>Learning agents:</strong> Learn from experience and adapt over time, using techniques like reinforcement learning to enhance their decision-making. An example is game-playing agents that improve through trial and error.</li><li><strong>Multi-agent systems:</strong> These systems feature multiple agents collaborating to solve problems, which is useful in robotics, traffic management, and supply chain optimisation.</li></ul><p><strong>AI agent use cases we likely encounter in the near future:</strong></p><p><strong>Gaming and Virtual Worlds</strong></p><ul><li>AI agents can make non-player characters (NPCs) behave more realistically and adapt to player actions, enhancing immersion.</li><li>Driving dynamic storylines that react to player choices and maintain a persistent, evolving game world even when not playing.</li><li>Agents could adjust game difficulty dynamically based on player skill level for an optimal challenge.</li></ul><p><strong>Customer Service and Sales. Next-level self-service (Health)</strong></p><ul><li>AI agents enable 24/7 personalised customer support, understanding needs through data analysis and automating routine queries.</li><li>They can tailor marketing messages, assist in sales processes, and elevate the overall customer experience.</li></ul><p><strong>Personal Productivity</strong></p><ul><li>Agents can act as intelligent assistants, understanding voice commands, scheduling appointments, sending messages, and retrieving information.</li><li>They can anticipate needs, provide recommendations, and automate repetitive tasks to boost productivity.</li></ul><p><strong>Content Creation and workflows</strong></p><ul><li>AI agents have the potential to assist with research, writing, and editing by understanding context and patterns.</li><li>They could enable collaborative writer-editor workflows with agents specialising in different roles.</li></ul><p>These use cases and AI agents have potential applications across customer service, healthcare, retail, education, and manufacturing industries. They’ll be able to automate responses, assist in medical diagnoses, personalise shopping and learning experiences, and optimise production schedules.</p><p><strong>How and what should we be considering when designing these types of interactions:</strong></p><ul><li><strong>Human Preferences and Needs:</strong> Ensure that humans are at the centre of the process when defining the problem space and solution that caters to their needs.</li><li><strong>Enable Natural Interactions (Agent Capabilities and Interactions): </strong>Define and enable humans to be able to interact with agents in a way that feels natural; this can be done through multi-model inputs.</li><li><strong>Provide Ethics, Transparency, and Control:</strong> Design clear onboarding to manage expectations and allow users to control agent decisions.</li><li><strong>Prioritise Simplicity and Efficiency (System Architecture and Scalability):</strong> streamline workflows to facilitate collaboration between human and AI agents, automate repetitive tasks, and surface relevant information to improve productivity.</li><li><strong>Embrace Iterative Improvement:</strong> Continuously gather user feedback and usage data to refine agent performance. Adopt a “build-measure-learn” approach.</li></ul><p>We are moving quickly, and if we thought moving from point-and-click (Desktop) to multi-touch (mobile) was fast, think again.</p><p><em>References:</em></p><ul><li>The Innovators by Walter Isaacson — <a href="https://en.wikipedia.org/wiki/The_Innovators_(book)">https://en.wikipedia.org/wiki/The_Innovators_(book)</a></li><li><em>Augmenting Human Intellect: A Conceptual Framework </em>— Douglas Engelbert — <a href="https://web.archive.org/web/20110504035147/http://www.dougengelbart.org/pubs/augment-3906.html">Link Web archive</a></li><li>The shape of AI — Emily Campbell — <a href="https://shapeofai.substack.com/">https://shapeofai.substack.com</a></li><li>SimForm — 6 types of AI agents <a href="https://www.simform.com/blog/types-of-ai-agents/">https://www.simform.com/blog/types-of-ai-agents/</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9cf2b0c2fd8a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Exploring the Convergence of AI and Design: Insights, Frameworks, and Tools for Success]]></title>
            <link>https://medium.com/@spl72/exploring-the-convergence-of-ai-and-design-insights-frameworks-and-tools-for-success-e998f8545e67?source=rss-faf3a54a5d03------2</link>
            <guid isPermaLink="false">https://medium.com/p/e998f8545e67</guid>
            <category><![CDATA[design]]></category>
            <category><![CDATA[design-principles]]></category>
            <category><![CDATA[design-patterns]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[design-process]]></category>
            <dc:creator><![CDATA[Justin Hevey]]></dc:creator>
            <pubDate>Tue, 26 Mar 2024 10:20:10 GMT</pubDate>
            <atom:updated>2024-03-26T10:20:10.826Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*F2d3drrPaWo2KhGUpHaqeA.png" /><figcaption>Midjourney</figcaption></figure><p>As a designer, it’s easy to feel overwhelmed by the rapid pace of change in our industry. With so many conversations in the design space right now around AI, leadership and design in general on the value it derives for business, what does this mean for design? I’m seeing and reading two different views: Excited to see what could be and what opportunities it unfold, or the fear of job loss and uncertainty about what the future holds as a designer and design industry at large.</p><p>With this in mind, I’ve embarked on a writing journey to delve into all things related to designing for AI. Through this, I aim to be curious, learn, understand, and grow as a designer.</p><p>I wanted to start by sharing foundations and areas I’ve explored in learning about designing with and for AI. These foundational topics have helped me find clarity and offered thought-provoking ideas and opportunities to explore where design may be heading in the next 2–3 years. Some brilliant designers are already providing excellent thought leadership in this space, thus providing optimism about the broader opportunities for design. It may feel like a headwind for design, which will depend on who you speak with, but this is a tailwind for design in my mind; this is exciting.</p><h3>Thought leadership</h3><p><a href="https://designingwithai.substack.com/"><strong>Designing with AI</strong></a><strong>.</strong> I love what Mia Blume has produced here with the community by sharing and opening up her learnings. She has had insightful conversations with designers pushing boundaries in AI exploration. Some of the highlights for me include conversations with the cofounders of <a href="https://designingwithai.substack.com/p/a-conversation-with-the-visual-electric">Visual Electric</a> and an interview with <a href="https://designingwithai.substack.com/p/john-maeda-defining-the-moment">John Maeda</a>. Highly recommend spending time over at Designing with AI.</p><p><a href="https://empathyandai.beehiiv.com/"><strong>Empathy and ai </strong></a>— Alex writes on Research, Design, and Humanity in the Age of AI, offering thought-provoking insights into potential directions for design, designers and AI. His discussions on AI agents, skills, and the concept of product as a partner are well worth exploring in terms of the opportunities for design and driving future customer value. In this <a href="https://empathyandai.beehiiv.com/p/reverse-reckoning-ux-design"><strong>article</strong></a><strong> —</strong><a href="https://empathyandai.beehiiv.com/p/reverse-reckoning-ux-design"><strong> The 3 Capabilities Designers Need for the AI Era</strong></a>, Alex outlines the three necessary skills for designers in this emerging field: strategy, interaction design, and understanding model design.</p><blockquote>We’ve moved from designing “waterslides,” where we focused on minimizing friction and ensuring fluid flow — to “wave pools,” where there is no clear path and every user engages in a unique way. — Empathy &amp; ai</blockquote><h3>Principals and Guidelines</h3><p><a href="https://pair.withgoogle.com/guidebook/">Google</a>, <a href="https://www.ibm.com/design/ai/">IBM</a> and <a href="https://www.microsoft.com/en-us/haxtoolkit/">Microsoft</a> have excellent guidelines, principles and frameworks for designing AI. Strongly recommended. IBM Design offers a free short course that I recommend. <a href="https://www.ibm.com/design/thinking/page/courses/AI_Essentials"><strong>design thinking to AI</strong></a><strong>.</strong></p><blockquote>AI isn’t about data, it’s about insights. Data is the fuel for the car, insights are the destinations. This is what people care about. — Fundamentals IBM Design</blockquote><h3>Frameworks</h3><p>Designers and product teams are developing, sharing and using numerous frameworks to design AI experiences. Recently, I’ve appreciated<strong> </strong><a href="https://medium.com/@sarahtan.twh/navigating-ai-design-the-new-blueprint-for-designers-8e41813d6acb"><strong>Sarah Tan’s medium series</strong></a> on developing a human-centred AI framework based on the five-step design thinking framework and a new blueprint for designers. I plan to adapt this framework in my next project. Sarah Tan’s five-step framework simplifies the process of understanding and designing with machines from a human perspective.</p><p><a href="https://aixdesign.co/toolkit"><strong>https://aixdesign.co/toolkit</strong></a> — offers designers and product innovators another framework and set of tools (Awesome prompt cards!) to aid in designing for advanced technology. Additionally, the community has a vast library of resources to explore.</p><blockquote>When tapping into AI innovation, the key question to ask is really — what AI capabilities can we leverage for disruptive innovation? —<a href="https://medium.com/@sarahtan.twh"> Sarah Tan</a> — formatif</blockquote><h3>Design patterns.</h3><p>Several sites are curating early AI design patterns. I believe as we continue to design and explore interacting with machines, we’ll quickly move beyond the chat interface, it will be interesting to see how these patterns evolve. The upcoming paradigm shift in interaction design promises exciting times with new forms of personalisation, customisation, and interaction with vast amounts of data.</p><p><a href="https://aiverse.design/"><strong>https://aiverse.design</strong></a> — good breakdown of patterns (conversational, assistive, invisible, ambient)</p><p><a href="https://www.notion.so/GenAi-Design-Patterns-78b958201bc54294aa952642c1ab72e1?pvs=21"><strong>New Design patterns empowering people</strong></a><strong> </strong>— exploring new AI patterns on trust, collaboration, data access and decision influence</p><p><a href="https://www.shapeof.ai/"><strong>https://www.shapeof.ai</strong></a> — a solid collection of evolving patterns in the AI space. I love how they have broken them into buckets, visual cues, iconography, trust, wayfinders, prompts and tuners. These patterns feel like the anatomy of the new paradigm shift.</p><blockquote>“We’re transitioning from designing pixels to designing patterns, from digital interfaces to experiences” Noah Levin, VP at Figma</blockquote><h3>Tools</h3><p>There are numerous resources available, making it difficult to review them all. Below, I’ve listed some tools that I believe are worth exploring on your AI journey.</p><p><a href="https://ai.boardofinnovation.com/"><strong>https://ai.boardofinnovation.com</strong></a> — The AI Toolbox for Innovators is an excellent design collaborator</p><p><a href="https://rationale.jina.ai/"><strong>https://rationale.jina.ai</strong></a><strong> </strong>— I think this is a really interesting tool and the power of AI. Helping business owners, managers, and individuals in making tough decisions.</p><p><a href="https://tome.app/"><strong>https://tome.app</strong></a> — Storytelling is such an essential skill for designers. Tome helps designers tell better stories by removing the mundane task of setting up slides and allowing the designer/presenter to focus on the narrative.</p><p>If you found this information helpful or wish to add to the discussion, please leave a comment. I would love to learn how you develop your AI skills and knowledge.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e998f8545e67" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Highlighting poor design]]></title>
            <link>https://medium.com/series/highlighting-poor-design-5c00a8864700?source=rss-faf3a54a5d03------2</link>
            <guid isPermaLink="false">https://medium.com/p/5c00a8864700</guid>
            <dc:creator><![CDATA[Justin Hevey]]></dc:creator>
            <pubDate>Sat, 13 Oct 2018 06:45:02 GMT</pubDate>
            <atom:updated>2018-10-21T08:55:23.664Z</atom:updated>
            <content:encoded><![CDATA[<p>Series highlighting poor design.</p><h4>Poor design #1. – The user should be able to do what the user wants to do.</h4><h4>Poor design #2. – The user should be able to do what the user wants to do.</h4><h4>I just want to sign up and hand over my money…</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5c00a8864700" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Crafting silverware in Cambodia]]></title>
            <link>https://medium.com/series/crafting-silverware-in-cambodia-bf85cfe35328?source=rss-faf3a54a5d03------2</link>
            <guid isPermaLink="false">https://medium.com/p/bf85cfe35328</guid>
            <dc:creator><![CDATA[Justin Hevey]]></dc:creator>
            <pubDate>Sat, 29 Jul 2017 13:55:53 GMT</pubDate>
            <atom:updated>2017-07-29T13:55:53.802Z</atom:updated>
            <content:encoded><![CDATA[<h4>Heating metal over coals using a air blower to control the heat to create the shape required.</h4><h4>Once they’re happy with the bowl’s shape. The design is finalised using a steel block and hammer to create the end shape and design of the bowl.</h4><h4>The bowl is then filled with a filler that allows the final design details to be crafted into the metal.</h4><h4>The final designs with detail added.</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bf85cfe35328" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Ocean Studies]]></title>
            <link>https://medium.com/series/ocean-studies-21bf0a1606bf?source=rss-faf3a54a5d03------2</link>
            <guid isPermaLink="false">https://medium.com/p/21bf0a1606bf</guid>
            <dc:creator><![CDATA[Justin Hevey]]></dc:creator>
            <pubDate>Sat, 25 Mar 2017 02:04:55 GMT</pubDate>
            <atom:updated>2017-03-25T02:04:55.975Z</atom:updated>
            <content:encoded><![CDATA[<img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=21bf0a1606bf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Dear Medium,]]></title>
            <link>https://medium.com/@spl72/dear-medium-21d229c79e88?source=rss-faf3a54a5d03------2</link>
            <guid isPermaLink="false">https://medium.com/p/21d229c79e88</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[bookmarks]]></category>
            <category><![CDATA[medium]]></category>
            <dc:creator><![CDATA[Justin Hevey]]></dc:creator>
            <pubDate>Sat, 17 Oct 2015 08:22:34 GMT</pubDate>
            <atom:updated>2015-10-17T08:43:28.934Z</atom:updated>
            <content:encoded><![CDATA[<p>Dear Medium,</p><p>Is bookmarks not a feature that’s required on the mobile native app?</p><p>It’s a constant frustration over the last number of releases that’s it out, in and now out again.</p><p>Please fix.</p><p>EDIT: wow I have searched deeper and found them at the bottom of search tab.</p><p>Interestingly why would you separate from my account/profile tab. If that’s where I go to see my stories and recommends, isn’t that where I would go to find my bookmarks too?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=21d229c79e88" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>