<?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 Aaron Sempf on Medium]]></title>
        <description><![CDATA[Stories by Aaron Sempf on Medium]]></description>
        <link>https://medium.com/@Aaron.Sempf?source=rss-522a7e208490------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*3ffLjELWHlrCG3nPJal9aA.jpeg</url>
            <title>Stories by Aaron Sempf on Medium</title>
            <link>https://medium.com/@Aaron.Sempf?source=rss-522a7e208490------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 04:21:39 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@Aaron.Sempf/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[Swarm Forming]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/c%D3%95lus-concept/swarm-forming-1842e85a9726?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1536/1*7oK1-McP_FRvRb_t7pt3hg.png" width="1536"></a></p><p class="medium-feed-snippet">How a drone becomes a member</p><p class="medium-feed-link"><a href="https://medium.com/c%D3%95lus-concept/swarm-forming-1842e85a9726?source=rss-522a7e208490------2">Continue reading on CӔLUS Concept »</a></p></div>]]></description>
            <link>https://medium.com/c%D3%95lus-concept/swarm-forming-1842e85a9726?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/1842e85a9726</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[drone-swarm]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[networking]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Sat, 29 Aug 2026 04:17:11 GMT</pubDate>
            <atom:updated>2026-08-29T04:17:11.884Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The Swarm-Mother]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/c%D3%95lus-concept/the-swarm-mother-9cb292b843a4?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1536/1*nng3YSxN7Ys507LpvR4aqQ.png" width="1536"></a></p><p class="medium-feed-snippet">Node 0: The node that belongs to two networks at once</p><p class="medium-feed-link"><a href="https://medium.com/c%D3%95lus-concept/the-swarm-mother-9cb292b843a4?source=rss-522a7e208490------2">Continue reading on CӔLUS Concept »</a></p></div>]]></description>
            <link>https://medium.com/c%D3%95lus-concept/the-swarm-mother-9cb292b843a4?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/9cb292b843a4</guid>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[drone-swarm]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Sat, 29 Aug 2026 04:10:32 GMT</pubDate>
            <atom:updated>2026-08-29T04:34:38.828Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Swarm Communication]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/c%D3%95lus-concept/swarm-communication-7eb79f17ba47?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1200/0*kUAN8yEcS5HXIeYI.png" width="1200"></a></p><p class="medium-feed-snippet">The intelligence layer floods, the tool layer routes, and both share one mesh</p><p class="medium-feed-link"><a href="https://medium.com/c%D3%95lus-concept/swarm-communication-7eb79f17ba47?source=rss-522a7e208490------2">Continue reading on CӔLUS Concept »</a></p></div>]]></description>
            <link>https://medium.com/c%D3%95lus-concept/swarm-communication-7eb79f17ba47?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/7eb79f17ba47</guid>
            <category><![CDATA[autonomous-agent]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[drones]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[communication]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Thu, 30 Apr 2026 01:56:58 GMT</pubDate>
            <atom:updated>2026-04-30T01:56:58.633Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Skill Sets]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/c%D3%95lus-concept/skill-sets-8154c9355281?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1200/0*CUCUHuNTdaDnyRkc.png" width="1200"></a></p><p class="medium-feed-snippet">What a drone knows it can do, and why that matters more than what it carries</p><p class="medium-feed-link"><a href="https://medium.com/c%D3%95lus-concept/skill-sets-8154c9355281?source=rss-522a7e208490------2">Continue reading on CӔLUS Concept »</a></p></div>]]></description>
            <link>https://medium.com/c%D3%95lus-concept/skill-sets-8154c9355281?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/8154c9355281</guid>
            <category><![CDATA[software-architecture]]></category>
            <category><![CDATA[drones]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[autonomy]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Mon, 20 Apr 2026 02:26:15 GMT</pubDate>
            <atom:updated>2026-04-20T02:26:15.345Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The Drone Brain]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/c%D3%95lus-concept/the-drone-brain-c6b6057eda3f?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1200/0*ubHze5aIIEFJi7Zf.png" width="1200"></a></p><p class="medium-feed-snippet">Three domains, three processors, one autonomous unit</p><p class="medium-feed-link"><a href="https://medium.com/c%D3%95lus-concept/the-drone-brain-c6b6057eda3f?source=rss-522a7e208490------2">Continue reading on CӔLUS Concept »</a></p></div>]]></description>
            <link>https://medium.com/c%D3%95lus-concept/the-drone-brain-c6b6057eda3f?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/c6b6057eda3f</guid>
            <category><![CDATA[autonomy]]></category>
            <category><![CDATA[swarm-intelligence]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Sat, 18 Apr 2026 02:25:34 GMT</pubDate>
            <atom:updated>2026-04-20T02:28:50.613Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[From Scale to Stability]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/agentic-ai-systems/from-scale-to-stability-f005d4f358a8?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1200/0*_SKj7pypiJD0qX6e.png" width="1200"></a></p><p class="medium-feed-snippet">For a long time, scale was the hard problem.</p><p class="medium-feed-link"><a href="https://medium.com/agentic-ai-systems/from-scale-to-stability-f005d4f358a8?source=rss-522a7e208490------2">Continue reading on Agentic AI Systems »</a></p></div>]]></description>
            <link>https://medium.com/agentic-ai-systems/from-scale-to-stability-f005d4f358a8?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/f005d4f358a8</guid>
            <category><![CDATA[autonomy]]></category>
            <category><![CDATA[ai-governance]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Sat, 18 Apr 2026 02:16:54 GMT</pubDate>
            <atom:updated>2026-04-18T02:16:54.626Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Why “Human-in-the-Loop” Is Not a Control Strategy]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/agentic-ai-systems/why-human-in-the-loop-is-not-a-control-strategy-f45fcebc1e6f?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1200/0*3LK7bniuPwNGLsW-.png" width="1200"></a></p><p class="medium-feed-snippet">When systems behave unpredictably, the instinctive response is to slow them down: add a review, insert a checkpoint, put a human back in&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/agentic-ai-systems/why-human-in-the-loop-is-not-a-control-strategy-f45fcebc1e6f?source=rss-522a7e208490------2">Continue reading on Agentic AI Systems »</a></p></div>]]></description>
            <link>https://medium.com/agentic-ai-systems/why-human-in-the-loop-is-not-a-control-strategy-f45fcebc1e6f?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/f45fcebc1e6f</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[automation]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[ai-governance]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Fri, 10 Apr 2026 12:43:41 GMT</pubDate>
            <atom:updated>2026-04-10T12:43:41.353Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The Quiet Collapse of Hierarchy]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/agentic-ai-systems/the-quiet-collapse-of-hierarchy-d4166ce19be2?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1200/0*bhPTcn0Fj6Ae2ylc.png" width="1200"></a></p><p class="medium-feed-snippet">Hierarchy did not fail dramatically.</p><p class="medium-feed-link"><a href="https://medium.com/agentic-ai-systems/the-quiet-collapse-of-hierarchy-d4166ce19be2?source=rss-522a7e208490------2">Continue reading on Agentic AI Systems »</a></p></div>]]></description>
            <link>https://medium.com/agentic-ai-systems/the-quiet-collapse-of-hierarchy-d4166ce19be2?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/d4166ce19be2</guid>
            <category><![CDATA[ai-governance]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[automation]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Mon, 30 Mar 2026 13:42:56 GMT</pubDate>
            <atom:updated>2026-03-30T13:45:18.148Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Autonomy Was Already Here]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/agentic-ai-systems/autonomy-was-already-here-467ff7448650?source=rss-522a7e208490------2"><img src="https://cdn-images-1.medium.com/max/1200/0*iLaSC7MrsGTi6Tv2.png" width="1200"></a></p><p class="medium-feed-snippet">Most conversations about AI assume autonomy is new.</p><p class="medium-feed-link"><a href="https://medium.com/agentic-ai-systems/autonomy-was-already-here-467ff7448650?source=rss-522a7e208490------2">Continue reading on Agentic AI Systems »</a></p></div>]]></description>
            <link>https://medium.com/agentic-ai-systems/autonomy-was-already-here-467ff7448650?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/467ff7448650</guid>
            <category><![CDATA[ai-governance]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[automation]]></category>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Mon, 30 Mar 2026 13:27:23 GMT</pubDate>
            <atom:updated>2026-03-30T13:51:17.446Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Development Uncertainty Principal]]></title>
            <link>https://medium.com/dev-managers-handbook/development-uncertainty-principal-1611d845774f?source=rss-522a7e208490------2</link>
            <guid isPermaLink="false">https://medium.com/p/1611d845774f</guid>
            <dc:creator><![CDATA[Aaron Sempf]]></dc:creator>
            <pubDate>Tue, 17 Mar 2026 06:28:20 GMT</pubDate>
            <atom:updated>2026-03-17T06:28:20.197Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*y4tiZhtX5waqcnTz2FMp3Q.jpeg" /><figcaption>Image designed by <a href="http://www.freepik.com/free-vector/confused-businessman_764462.htm">Freepik</a></figcaption></figure><p>One of the greatest things I ever learned, was how to learn… How to diagnose problems, to ideate solutions, to test-fail and retrospectively learn from success or failure. But of all the techniques and processes for learning, in my opinion, the best tool at our disposal is observation.</p><p>Now unless you&#39;ve studied science or psychology in some shape or form, you’re probably thinking, “well of course, I know how to observe. I watch what goes on around me” — Granted, watching is a big part of observation, but observation in essence is inspection and scrutinisation beyond what happens in front of us. Not just watching the result of an action, but finding the cause of that action through scrutinising and learning what lead to that action.</p><p>Observation over time within different environments, in my career has been invaluable. Particularly as a Technical Analyst who needs to understand a given environment or workflow in order to improve or automate that environment, but also just as importantly, if not more, the ability to observe ones own and ones teams practices and culture at each level.</p><p>As an exercise in learning through observation, for the past year instead of being the less than humble know-it-all, pushing my applied knowledge of best practices and standards for development and project management, or competing with ideas and architecture for the best solution, I have spent the time actively taking a step back and observing, within my team and organisation and through discussion of similar environments within my circles.</p><p>What started as a lesson in humility, turned into a quest to learn from those around me what fruit would bear without my ‘interference’… and for those of you who HAVE studied science or psychology in some form, lets not discuss <a href="https://en.wikipedia.org/wiki/Schr%C3%B6dinger%27s_cat">Schrödinger’s Cat</a> of this situation, as I was a member of the team prior to ‘observing’ the team, so what came out of my observation would have already been there having me observed it or not.</p><p>And what I learned, stated at the top, — what undoubtedly caught your attention to read this article — is that while there are <a href="https://en.wikipedia.org/wiki/List_of_software_development_philosophies">many different practices and methodologies</a> that all solve a particular issue within the development and project management life cycle, none that I have observed have perfectly or seamlessly accounted for all aspects of the environment in which they are applied.</p><p>But all reveal the same unaddressed aspects in the development life cycle, as well as in many cases the environment itself, for which I call the Development Uncertainty Principal.</p><p><em>— I may have named it loosely after </em><a href="https://en.wikipedia.org/wiki/Uncertainty_principle"><em>Heisenberg’s uncertainty principle</em></a><em>, for the way in which it reveals that the more closely one pins down one aspect of development, the less precise another aspect becomes, but I&#39;m not sure. —</em></p><p>The Development Uncertainty Principal is the level of uncertainty around any aspect of development or the management of a development environment that directly effects the development, the people, or the environment.</p><p>So, right now if you are saying, “well, there wont be any uncertainty if you define everything”, than you’re either a Business Analyst or you’ve never been a developer or both.</p><p>No matter how much you define, there will always be some level of unknown, some uncertainty.</p><p>For example, when breaking a development project down into its components or features, or whatever your chosen methodology calls the modularised portions of development, there are 2 categories, no matter what methodology you use, that each component can be grouped into; things that have been done before, and things that have not been done before.</p><p>If it has been done before, copy it, reuse it, buy it, hire the person who did it, whatever it is, because as developers we don’t need to be solving problems that have already been solved, the only problems we want to be solving ourselves are the ones where there is something new, something that we’ve never done before as part of that component.</p><p>If we are doing something that we have never done before, it carries with it a small amount of risk. It might not work, or it might take a few iterations to get it working correctly.</p><p>We can’t be absolutely certain how long its going to take, there is an ‘uncertainty’ to the amount of time it will take to develop the component.</p><p>On top of the known uncertainty of developing things we haven’t developed before, there are also the things we don’t know that we don’t know, that no amount of analysis will reveal.</p><blockquote><a href="https://theitriskmanager.wordpress.com/2015/05/10/understanding-uncertaintys-impact-on-learning/">People would prefer to be wrong than uncertain</a>. <br>— Chris Matts</blockquote><p>Unfortunately, as humans, we don’t like uncertainty.</p><p>Especially from a project managers perspective, we hate uncertainty (<a href="https://en.wikipedia.org/wiki/Confirmation_bias">confirmation bias</a>). we cant go to the client and say, “it might take this long, but it could be longer or shorter, we really don’t know”. but the hard truth is that people would rather lie, or hear a lie than be uncertain.</p><p>And we make every effort to eliminate it.</p><p>However, the most common analysis practice of breaking down the problem into its components, for new things or things we don’t fully understand yet, doesn’t eliminate the uncertainty, it hides it.</p><p>When we try to break something that we don’t understand into smaller components, we are trying to define or add certainty to that which we may not yet have all the information for. It’s only when we start developing that we start to understand what we didn’t know, and the uncertainty is again revealed.</p><p>But the uncertainty of estimation is merely one area where we can easily find uncertainty</p><p>uncertain acceptance criteria</p><p>with behaviour driven development, when talking through all the scenarios and you find a context that no one knows the outcome to, you have uncovered some uncertainty, the something new, the something that we’ve never done before.</p><p>Nothing is ever “Finished” once its through the SDLC even in production, there still may be bugs, or the way users interact with it may be unexpected and stimulate changes, or it may have been completely wrong and need to build something new. It needs to be maintained.</p><p>Until the software is removed from production, and no longer required, it is never finished, its in an endless state of awaiting validation.. sometimes it may take months or many months to receive validation, but it is always in a state of awaiting validation as long as someone is still using it.</p><p>uncertain team culture, ego vs learn</p><h4>Using the Uncertainty Effect</h4><p>Estimations:</p><p>Estimations are inaccurate due to the ‘uncertainty principal’.</p><p>Break the estimations into 2 categories, Expectations, where we can expect something will take an amount of time due to having done it before or the time it takes to buy, etc. and Hopes, where we hope it will take an amount of time for all the things in it that we can see, but we are uncertain of things that may come up during development.</p><p>Acceptance Criteria:</p><p>instead of insisting that they go and think about it, collaborate, try things, explore… making sure that its something that is easy to change, because it will change.</p><blockquote>Don’t pick the right technology. Pick the technology that’s easy to change.<br>- Chris Matts</blockquote><p>You dont know what the right technology is, otherwise you wouldn’t be having to make a choice. Pick a technology thats easy to change. It might be wrong, but when more information comes later you’ll know what to change to. (or keep if it turned out to be right).</p><p><em>Whenever we have a choice and not enough information to help us make the decision, picking a decision that’s cheap to change later usually turns out to be a good way forward, as we can change when we have more information (or keep it if it turns out to be good enough!)</em></p><p>- Finding the Uncertainty in the team.</p><p>Nurture honesty and security in the team</p><p>Its easy to suggest that a culture of honesty and security in an organisation, is driven from the top, or comes from the manager to set the environment for the team… but what if your the manager, what if you are at the top, then this</p><p>Dont let your process hide your ignorance… discovery of information, find the truth of the situation</p><p>conclusion:</p><p>Development form inital concept to users and maintenance is not a series of serial steps. it should be an interactive process involving all proponenets of the project at all steps of the project. ie, getting your users involved as early as possible… Human centred design</p><p>references:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F6lONh2bHHjI%3Ffeature%3Doembed%26start%3D0&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D6lONh2bHHjI&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6lONh2bHHjI%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/95c3e942bcf7ffbbab7aa5dad702a826/href">https://medium.com/media/95c3e942bcf7ffbbab7aa5dad702a826/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1611d845774f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/dev-managers-handbook/development-uncertainty-principal-1611d845774f">Development Uncertainty Principal</a> was originally published in <a href="https://medium.com/dev-managers-handbook">Dev Managers Handbook</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>