<?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 Eleahminer on Medium]]></title>
        <description><![CDATA[Stories by Eleahminer on Medium]]></description>
        <link>https://medium.com/@eleahminer?source=rss-c65edfb60b19------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*trU_cMnk5jj96EkF</url>
            <title>Stories by Eleahminer on Medium</title>
            <link>https://medium.com/@eleahminer?source=rss-c65edfb60b19------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 22:51:41 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@eleahminer/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[How I Use AI at Home as a Mom]]></title>
            <link>https://medium.com/@eleahminer/how-i-use-ai-at-home-as-a-mom-e1b612c3a56a?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/e1b612c3a56a</guid>
            <category><![CDATA[mental-health]]></category>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[parenting]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[motherhood]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Thu, 24 Sep 2026 13:01:02 GMT</pubDate>
            <atom:updated>2026-09-24T13:01:02.908Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*SQjib9pVtccPqwa6.jpg" /></figure><h4>Seven things I hand to AI at home, with the prompts so you can copy them.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*hg0uFGv1bOtSQZt0.jpg" /></figure><p>So last week I gave a talk in front of over 200 people, mostly men, about how I use AI in my business. And somebody asked me how I would describe the difference between how men and women create.</p><p>I said if I had to put it into one word, it would be flexibility. We are multitasking a million things at once. Most of us don’t have eight hours a day to time block and work. Moms especially. I have had all of those time constraints myself, and I still do. Someone recorded the whole thing, <a href="https://www.youtube.com/watch?v=Bl9-3AHeW-U">it’s here</a> if you want to watch it.</p><p>And I think that is why so much of how I use AI has nothing to do with work. I have an agent on my team named Flint, and his whole job is my life. School, health, money, food. TikTok has started calling this a digital co-parent. Mine reads the school email and puts the dates on my calendar, which is a smaller job with a smaller name. But it takes a real piece of the mental load off, and I don’t hear many people talking about it.</p><p>I’ll go deep on a few and then rattle off the rest.</p><h3>Processing my therapy</h3><p>This is how I got into AI in the first place. People ask me that a lot, and the answer is not a business answer.</p><p>I have an amazing therapist, and I have been in therapy pretty heavily for about three years. I grew up in a fundamentalist Christian cult, and there is a lot to rewire. A quick example is self-care. The way I was raised, taking care of yourself was selfish, especially as a woman. You serve men, you serve other people, you serve your kids, and you come absolutely last. I am still working on that one. Things like that take time to integrate.</p><p>My therapist is still my therapist. We all know why AI shouldn’t be one. It will tell you what you want to hear. What I wanted was help with the six days in between sessions.</p><p>So I asked her if I could record and transcribe our sessions, and she said yes. I now have at least a year of them saved. I can look back and see how I’ve grown, which is kind of wild to read. And I can take what came up on Tuesday and keep working with it through the week.</p><p>The way I started doing that was through somatic journeys, which are basically guided meditations. I had just finished an inner voice facilitator training, and my therapist is a certified somatic facilitator too, so I would tell her what I was learning and she would say, yeah, that is what I learned in school. I would take a theme from a session and have my team help me shape a journey around it. Quieting the mind. Feeling an emotion physically in the body. Holding fear and joy at the same time and letting my nervous system know that is okay.</p><p>For a while I led a free one every week for my community, and a few of them are on my podcast. <a href="https://podcasters.spotify.com/pod/show/lea-miner/episodes/S2E6-A-Somatic-Journey-for-Finding-Peace-in-the-Middle-of-Chaos-e3ojr3k">A Somatic Journey for Finding Peace in the Middle of Chaos</a> and <a href="https://podcasters.spotify.com/pod/show/lea-miner/episodes/S2E3-A-Somatic-Journey-for-When-Youre-Overstimulated-e3nmcrv">A Somatic Journey for When You’re Overstimulated</a> both came out of this.</p><p>Someone told me this is a very meta way to use AI. Maybe. But if I can make my nervous system a better place, I think it affects everybody around me.</p><p><strong>If you want to try it.</strong> Ask your therapist first. Mine said yes, and some will say no, and that is theirs to decide. Then after a session, paste in the transcript or just your own notes and say:</p><blockquote><em>Here are my notes from therapy this week. Don’t give me advice and don’t analyze me. Pull out the two or three themes that came up, in my own words. Then write me a five-minute guided practice I can do on Thursday that helps me feel one of those themes in my body. Keep it gentle.</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*1MsPqJALYaPdpsFd.jpg" /></figure><h3>Parenting</h3><p>I want to keep this part vague. My son and I have a little agent I’ve set up as Aunt Lumen. When he wants a really hilarious story told to him, we ask her. When he’s struggling with something, I sit next to him and we talk to Aunt Lumen together, and she helps me explain a thing that feels like common sense to his brain.</p><p>We’ve made meditations together too. He picks a snake and an emotion, and we practice the breathing. It is called <a href="https://open.spotify.com/show/778YNm2wiSZK5Wk3NaEeoY">Super Snakes That Meditate</a>. It is not a high production value podcast. But I think what people want from a podcast lately is realness, and he is very real.</p><p><strong>If you want to try it.</strong> Sit next to your kid, not across the room, and say:</p><blockquote><em>Tell my kid a funny two-minute story about [the thing he loves right now] where the main character figures out [the thing I’m trying to explain]. No lesson at the end. Stop twice and let him pick what happens next.</em></blockquote><h3>School emails and the calendar</h3><p>It is so hard being a co-parent when you don’t always know what happened at school.</p><p>So any email that comes from school, Flint saves into one running file of what is going on there. Dates go on my calendar. I am no longer forgetting about the random days when there is no school, and it feels like there are a lot of those.</p><p>That is the whole thing. It took no special skill to set up. It just means the information is in one place when I need it, and I show up to conversations about my son knowing what is going on.</p><p><strong>If you want to try it.</strong> Forward or paste the school newsletter into Claude or ChatGPT and say:</p><blockquote><em>This is this week’s email from my kid’s school. List every date, deadline, and thing I need to bring or sign. Put each one in a format I can add to my calendar, with a reminder the night before. Then give me a three-line summary I can keep in a running note called School.</em></blockquote><p>If your AI is connected to your calendar, it can add them for you. If not, it takes two minutes to tap them in.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*TKLFO24_62tB0fFT.jpg" /></figure><h3>The pediatrician and the healthcare system</h3><p>Last week I took my son to the pediatrician. They talk fast. There is all sorts of information coming at you, and you are also parenting in that room.</p><p>So I started a voice note on my phone. When I got home I had my team clean it up, remind me exactly what was said, and hold on to it so I don’t forget to schedule the referral we were given. All the details and the resources his pediatrician mentioned were there when I went looking.</p><p>It lets me fiercely advocate for my son. I am really grateful for that.</p><p>I’ll say this part too. Getting on Medicaid in the first place was so hard. I think most moms who are struggling financially run into this. The system makes it hard to figure out where to start or how to apply, and on top of that I felt a decent amount of shame. I can’t believe I’m in this place. But having Medicaid for me and my son has been life-changing. I can get an urgent care appointment when we need one. If you know a woman who is stuck there, send her my way. I would love to help.</p><p><strong>If you want to try it.</strong> Ask the doctor if you can record so you can remember everything. Then:</p><blockquote><em>This is a voice note from my child’s doctor appointment. Clean it up and give me: what the doctor said, in plain words. Every follow-up I need to schedule, with who to call. Any medication or dose mentioned, exactly as it was said. Questions I should ask next time. Flag anything you are not sure you heard right so I can check it with the office.</em></blockquote><p>That last line matters. It is a transcript, and it can mishear. Anything medical, I confirm with the office.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*AjL7OtLTkbr9cnuo.jpg" /></figure><h3>The rest, quickly</h3><p><strong>Protecting my nervous system.</strong> There is one person whose emails I know will activate me. So my team reads those first and helps me figure out how to respond. I can always go read the original. I am still making every decision. I just don’t need to take every message in with my own eyes the second it lands. I have a client who does this now. When she gets a stressful email from a coworker, she has Claude read it first and help her reply.</p><p><strong>Taxes.</strong> I have an LLC, and taxes always felt daunting. I use TurboTax, and I have Claude look over my shoulder as a sounding board while I go. It is a second set of eyes, and the filing is still mine.</p><p><strong>Cooking.</strong> I love to cook. My favorite thing is taking whatever is left in the fridge and making something I’ve never made. Now I describe what sounds good, take a picture of the fridge and the pantry, and get a recipe that walks me through it like I have a private chef.</p><p>Around the 4th of July I wanted to recreate the Date Night Burger from Southern Sun here in Colorado. Dates, jalapeño, bacon, goat cheese. I described it, got a recipe, and then made it mine. Brioche pretzel buns from Trader Joe’s, sweet chili sauce, and mint and basil from my little herb garden. I think mine was better, for me anyway.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3KCF-Y8eVLt7i-cu.jpg" /></figure><blockquote><em>Here is a photo of my fridge and my pantry. I’m craving something creamy and a little spicy, and I have 40 minutes. Give me one recipe using mostly what I have, tell me the one or two things worth buying, and walk me through it step by step.</em></blockquote><h3>Where I’d start</h3><p>Pick the one that made your shoulders drop when you read it. For most moms I talk to, it is the school email one. Do it once this week and see how it feels.</p><p>And if you want help setting any of this up for your own life, I do free fifteen-minute calls, which are really just a question and a next step. <a href="https://leala.co/consult">Grab one here</a>. I would love to hear which one you tried.</p><p><em>This is part of an ongoing series on building a life and a business with AI as a non-technical founder. The last one, </em><a href="https://leala.co/articles/the-bottleneck-was-never-the-ai"><em>The Bottleneck Was Never the AI</em></a><em>, is about why AI tools stall inside a business. This one is the home version.</em></p><p><em>I’m building Leala, a somatic healing and inner voice community. Everything I’m learning about AI, nervous system regulation, and creating from alignment gets shared in real time.</em></p><p><em>If something in this piece landed, let’s talk. </em><a href="https://leala.co/consult"><em>Book a free fifteen minutes →</em></a></p><p><a href="https://www.linkedin.com/in/eleah-miner/"><em>LinkedIn</em></a><em> · </em><a href="https://www.instagram.com/leala.io/"><em>Instagram</em></a><em> · </em><a href="https://www.youtube.com/@leala.io"><em>YouTube</em></a><em> · </em><a href="https://leala.co/"><em>leala.co</em></a></p><p><em>Originally published at </em><a href="https://leala.co/articles/how-i-use-ai-at-home-as-a-mom"><em>leala.co</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e1b612c3a56a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Systems Built to Run Without Me]]></title>
            <link>https://medium.com/@eleahminer/systems-built-to-run-without-me-bcedaffe4ee3?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/bcedaffe4ee3</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Tue, 08 Sep 2026 23:46:15 GMT</pubDate>
            <atom:updated>2026-09-08T23:46:15.497Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3thpWjk2WuTdPec9.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*hlGdQyD44hDDSYa3.jpg" /></figure><p>So I had seven minutes on a Thursday night in Boulder, and twelve slides, and I was nervous.</p><p>This is that talk, written out. I have added the two pieces there was no room for on stage. One is what “solve the chaos first” looks like with a real client, because on the night it was a single line. The other is what happened after I stepped off the stage, which turned out to be the part I keep thinking about.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*6h3IfVUs_Bkv-uPO.jpg" /></figure><h3>Where I learned this.</h3><p>The first job I ever had, in college, they put me in charge of the billing department and the marketing department of a home security company. The owner would just say, “Leah, you’re the boss. Figure it out.” And he would hand me stacks of books. Good to Great. Seven Habits. How to Win Friends and Influence People. Kind of before my time, but the principles hold.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VHZaDVQgrZfZWHUm.jpg" /></figure><p>If a company depends on one person, it is fragile. You teach people to run the business when you are not there.</p><p>The one that stuck, from Jim Collins</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3KLuVceQNmnJGO0F.jpg" /></figure><h3>The problem I had to solve.</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GOWT8MSnjfUabgUQ.jpg" /></figure><p>If a system can support a family of five on twenty hours a week, then the weeks I have more time, it compounds.</p><p>I’m a solo mom and the primary breadwinner for a family of five. I have about twenty hours a week that I know I can work, and those hours have to be flexible, because at any moment I could get a call from my son’s school. I pause, and I pick it up later.</p><p>At first that felt impossible.</p><p>But lately my systems have started to work. I’m teaching pieces of my business to run without me. When somebody in my house has a sick day, my projects still move. Instead of seeing that as a weakness, I’m starting to see it as a superpower. If a system can support a family of five on twenty hours a week, then the weeks I have more time, it compounds.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*aGXY6xmhXg-5oMhs.jpg" /></figure><h3>Solve the chaos first.</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*pnSHmG_hxKcRixUN.jpg" /></figure><p>You cannot solve your other problems until you solve the chaos. If your nervous system is activated, your work is going to be far worse than when you are calm.</p><p>A lot of my clients have brains like mine. Almost every one of them feels like they are in chaos right now.</p><p>So I build slowly, on purpose. I go after whatever is causing the chaos first. Most of the time the biggest stressor is money, so we go after the revenue-driving work first. Sometimes it is communication. Sometimes it is needing to hire the right person.</p><p>Everything we do has a process, and each piece gets one percent better every time I touch it. I took that from James Clear. You cannot solve your other problems until you solve the chaos. If your nervous system is activated, your work is going to be far worse than when you are calm.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*P8IRFWOlS1EM5Vl2.jpg" /></figure><h3>What that looked like with one client</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CrF8iCNu-zG9WHIV.jpg" /></figure><p>The part I had no time for on stage</p><p>A founder came to me wanting a command center. One dashboard with her email stats, her site analytics, her membership numbers by tier, and her team’s hours, all in one place. It was a good idea, I think. I wrote the proposal, built it as a live page in her brand, and priced it.</p><p>Then on the call she told me the dashboard was a pet dream.</p><p>What was actually happening underneath: she had spent her savings on her team that year. She was loaning her own salary back to the business. She had paid one contractor around five thousand dollars for maybe a thousand dollars of load-bearing work. And she had no visibility into anyone’s hours, which is exactly why she wanted the dashboard in the first place. But the dashboard was weeks of building away from the leak.</p><p>So we moved. The first thing I’m doing for her is the broken-things list on her platform, priced at the same hourly rate her developers charge. I’m reviewing it from her actual code rather than from the titles on her tickets. The second is a consolidation she did the math on herself, out loud, while we were talking. Three subscription tools plus a person moving recordings by hand. About a thousand dollars a year, plus the hours.</p><p>The dashboard is still on the list. It is just not first.</p><p>That is what going after the chaos means in practice, and it is usually less fun than it sounds. Most of the time it means telling a client that the pretty thing waits.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*A_ntA6rlclb3uxpc.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Sfyi3eb4GcpM4N67.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*h2eg09Q_ZLmWGrWz.jpg" /></figure><p>The two files, as a skeleton</p><p>This is the shape of the two files everything else hangs off. Make the master prompt first, then one active context file per client. Keep them short enough that you will actually update them. If a section feels hard to write, that is the section your team has been guessing at.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NZTZDGg_Cvm2WjFw.jpg" /></figure><p>00_Master_Prompt.md · 00_Active_Context.md</p><pre><br>00_Master_Prompt.md<br># Who I am<br># What I do, and who I do it for<br># My clients, one line each<br># Rules I will not break<br># How I speak, and what I never say<br># What &quot;done&quot; looks like to me<br><br>Clients/&lt;Name&gt;/00_Active_Context.md<br># Where we are right now<br># What they asked for, in their words<br># Decisions already made, dated<br># Open questions I owe them<br># What I am not doing yet, and why<br># Transcripts: ./transcripts/YYYY-MM-DD.md</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ERQSQVY1cYrgqm7C.jpg" /></figure><p>Seven on the executive team, and I am the only one that is human.</p><p>And then there is my team. I always tell people I have seven on my executive team, and I’m the only one that is human.</p><p>Lumen is my chief integration officer and oversees everyone. Sable does design. Sol runs revenue and clients. Wren does content and infrastructure. Soren handles my voice notes, and Atlas is my Claude Code engineer, the one I have been working with most lately.</p><p>Inside each process, I have set up skills that teach the team exactly how I want something done. And I manage them by talking. I dictate from the gym, on a walk between calls, in the car. Every morning I do my morning pages and give the team feedback.</p><p>I’m kind of all in on letting AI support what I’m trying to do in my life, because I really need it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DwiRpwIyqW01UOKK.jpg" /></figure><h3>What happened after.</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*QNY32km4IZBUDT5N.jpg" /></figure><p>The exact quality that got me passed over for jobs was the thing people lined up to talk about.</p><p>I said thank you and stepped off, and then there was a line of people.</p><p>People wanting to do work with me. Women wanting to connect and collaborate. A lot of people asking how I made the slides, and how I made the website, and what skills I used. Which was funny, because my honest answer is that a lot of it is my vision for things.</p><p>One woman said I had put into words a lot of the things she had been thinking about and could not find the words for. And then she thanked me. I’m still a little bit undone by that one.</p><p>The thing I keep coming back to is older than the talk. For years I tried to be myself and got rejected for jobs over it. I think people could always just see it: oh, you’re an entrepreneur, and we want you to do this one narrow thing. I kind of would have loved a boring job for some time. And then I stood in a room in Boulder. The exact quality that got me passed over was the thing people lined up to talk about.</p><p>I was not wearing my recorder, which I regret, so this is the version I have.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3FsSO_hLah6neGrm.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*bqMMep0BnPfZQaXk.jpg" /></figure><p>If you want help building yours</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*10i_z5Hy7WIgt-fT.jpg" /></figure><p>I have a few spots open this month.</p><p><a href="https://calendar.app.google/yCAqyZvEEmKfogXG9">Book a time →</a></p><p><strong>If you want help building yours, I’m at leala.co/book.</strong></p><p><em>Originally published at </em><a href="https://leala.co/articles/systems-built-to-run-without-me.html"><em>https://leala.co</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bcedaffe4ee3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Bottleneck Was Never the AI]]></title>
            <link>https://medium.com/@eleahminer/the-bottleneck-was-never-the-ai-6cadd2580ab9?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/6cadd2580ab9</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[leadership]]></category>
            <category><![CDATA[small-business]]></category>
            <category><![CDATA[self-improvement]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Thu, 27 Aug 2026 16:14:03 GMT</pubDate>
            <atom:updated>2026-08-27T16:14:03.909Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gStSKVpJxDj-UdK8Qq33GA.png" /></figure><p>A client of mine has a coaching agent, a voice model, and a full content system already built. She still can’t get away from her desk.</p><p>Every piece of content that goes out still routes through her twice. Once to get assigned. Once to get rewritten, because whoever wrote the first draft couldn’t sound like her. She has the tools. She built the tools with me. And she is exactly as bottlenecked as she was before any of it existed, just with more software open.</p><p>I sat with that for a while before I said anything to her. Because my first instinct was to think something was wrong with the tools. It wasn’t. The tools were fine. What was underneath them wasn’t.</p><h3>What the research keeps confirming</h3><p>I went looking for whether this was just her, or whether it was a pattern. It’s not just her.</p><p>MIT’s Project NANDA found that 95% of generative AI pilots inside companies produce no measurable return, even with $30 to $40 billion in enterprise spend behind them. Not a modeling problem. Their own language for it is a learning gap: the tools don’t retain what they’re taught, they don’t adapt from one interaction to the next, and most of all, they don’t integrate into how the business actually runs. One manufacturing executive they interviewed put it plainly: the hype on LinkedIn says everything has changed, but in his operations, nothing fundamental has shifted.</p><p>PwC surveyed over four thousand CEOs and got a similar number from a different angle. Fifty-six percent saw zero cost or revenue improvement from AI. The pattern PwC named was deploy first, ask what problem you’re solving second.</p><p>Terminal X went a layer deeper, and this is the piece that actually reframed it for me. They found that companies need three things in place before AI gets deployed, not after: a way to measure whether anything worked, the infrastructure for the tool to actually connect to real workflows, and a strategy that treats the rollout as something the whole org has to metabolize, not just install. Skip any of those three and it doesn’t matter how good the model is.</p><p>And the most recent number, from a McKinsey survey published two days ago as I’m writing this: only 37% of companies can attribute any bottom-line impact to AI at all, flat since last year. Only 6% count as real high performers. McKinsey’s explanation for the gap between that 6% and everyone else wasn’t tool quality. It was organizational change. The 6% redesigned how decisions get made and who owns what. Everyone else added a tool to the org chart they already had.</p><p>That’s six independent reports, four different research groups, two different years, all landing on some version of the same sentence: the system it’s landing on determines whether AI does anything at all.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2kfg4f-mNJS5szFRz8FgVQ.png" /></figure><h3>Two ways a system can be broken</h3><p>Once I had language for this, I started checking it against actual clients instead of just my one example. And what I found is that “broken system” isn’t one thing. It’s at least two, and they need completely different fixes.</p><p>Shape one is the bottleneck. This is my client above. Everything has to pass through one person because nothing downstream can act without her. There’s no single place a decision lives once it’s made, so it lives in whoever remembers making it. Nobody besides her has standing to move a piece of work forward without her sign-off. And “sounding like her” only exists in her head, so every AI output is still a first draft that has to come back to her to become real. Add AI to that shape and you haven’t removed the bottleneck. You’ve given it more to check.</p><p>Shape two is the blind spot. I saw this with a different client, a small coaching-certification business. Their website broke four separate times over several months, including an integration that silently rerouted every new lead to the founder personally instead of the round-robin it was supposed to hit. There was a named owner for the website. He’d taken full responsibility for it. Ownership wasn’t the problem. Detection was. Nobody found out for months, because nothing was checking whether the funnel actually worked. The founder discovered it the way most people discover a blind spot: she noticed the effect, a sudden cash crunch, months after the actual cause. AI can’t fix a hole nobody can see, and neither can a new hire, or a better tool, or more spend.</p><p>Both of these get called “our systems are broken.” Both make new AI investment a waste until the specific missing piece gets installed. But they’re not the same missing piece.</p><p>Before I recommend any AI tool to a client, I check for which shape they’re in first. Sometimes it’s both.</p><p>For the bottleneck shape: Is there one place a decision lives once it’s made, or does it live in whoever remembers it? Does anyone besides the founder have standing to move work forward without her sign-off on every draft? Does the standard, the thing that makes it sound and look like them, exist as something an AI tool could actually be pointed at, or does it only exist in their head?</p><p>For the blind spot shape: Is there a baseline and a check that would surface a failure within days, or does it only surface once the damage has already piled up? If something is working, can you point to the number that proves it, or is “it seems fine” the whole measurement?</p><p>None of these questions are about AI. That’s sort of the point. AI amplifies whatever is already true about how the business runs. Layer it onto a bottleneck and the founder becomes the bottleneck faster. Layer it onto a blind spot and it fails the exact same silent way the last three tools already did.</p><h3>The part I haven’t fully resolved</h3><p>Here’s something I’m still sitting with, and I’d rather say it out loud than pretend I’ve got it figured out.</p><p>Clients almost always want to hire before the system underneath is solid. A social media person, an ops hire, someone to take something off their plate. My instinct is to tell them to pump the brakes and fix the system first. But that instinct makes me nervous too, because it puts me in the position of being the one holding everything until it’s ready, instead of them just going and getting the help. I don’t fully trust that a new hire walks into a broken system and doesn’t just become part of the breakage. I also know I could train one.</p><p>There’s research that backs the brakes instinct even outside of AI specifically. Startup Genome studied over three thousand high-growth companies and found that 74% of startup failures trace back to premature scaling, meaning the company invested in the next stage, more spend, more hires, more surface area, before the current stage was actually solid. The ones that sequenced it correctly grew roughly twenty times faster than the ones that didn’t. That’s not overcaution. That’s the single most correlated factor in whether a company survives its own growth.</p><p>So maybe the reframe is: I’m not competing with the future hire. I’m the reason the hire succeeds instead of inheriting a mess. I’m not fully settled on that yet. I wanted to include it anyway, because it’s the same pattern from the inside, and I think naming the tension is more honest than pretending I arrived somewhere clean.</p><h3>Where this actually goes</h3><p>If you’re staring at an AI tool that isn’t doing what you hoped, the question probably isn’t which tool to try next. It’s which of these two shapes your business is actually in, and whether anyone has looked.</p><p>I’ve started calling this a systems check. It’s really just an hour where I walk through the five questions above against how your business actually runs today, not how you’d describe it on a good day. Most people leave it with a clear answer to something they’d been vaguely worried about for months.</p><p>If that sounds like where you are, I have a few spots open this month.<a href="https://calendar.app.google/yCAqyZvEEmKfogXG9"> Book a time →</a></p><p>The tools were never the problem. The question is just whether anyone checked what they were landing on.</p><p>⸻</p><p><em>This is the third in a series about building with AI as a non-technical founder. The last one,</em><a href="https://medium.com/@eleahminer/the-master-prompt-whats-actually-in-mine-1b1782a18eab"><em> The Master Prompt: What’s Actually in Mine</em></a><em>, covers how I built the document that tells AI who I am. This one covers what happens after the tools are built.</em></p><p><em>I’m building LEALA, a somatic healing and inner voice community. Everything I’m learning about AI, nervous system regulation, and creating from alignment gets shared in real time.</em></p><p><em>If something in this piece landed, let’s talk.</em><a href="https://calendar.app.google/yCAqyZvEEmKfogXG9"><em> Book a coffee →</em></a></p><p><a href="https://www.linkedin.com/in/eleah-miner-844969b8/"><em>LinkedIn</em></a><em> · </em><a href="https://www.instagram.com/leala.io"><em>Instagram</em></a><em> · </em><a href="https://www.youtube.com/@leala-expansion"><em>YouTube</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6cadd2580ab9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What I Put in My Master Prompt]]></title>
            <link>https://medium.com/@eleahminer/the-master-prompt-whats-actually-in-mine-1b1782a18eab?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/1b1782a18eab</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[productivity]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Mon, 24 Aug 2026 15:41:01 GMT</pubDate>
            <atom:updated>2026-08-24T18:14:46.785Z</atom:updated>
            <content:encoded><![CDATA[<h3>Why you would want one</h3><p>People kept asking to see the real thing instead of hearing me describe it, so here it is. Structure first, then what goes in each part.</p><p>A master prompt is not a prompt. It’s a document about you that every AI conversation inherits, so you stop introducing yourself from scratch ten times a day.</p><p>Here is why that matters. When AI hands you generic advice, it is not because the tool is stupid and it is not because you are. It is because the tool knows nothing about you, so it gives you the average of everyone. Writing this down once ends that. It takes about an hour.</p><h3>What’s in mine, section by section.</h3><p><strong>1. The person.</strong> Where I live, who’s in my house, my kid, my partner, my health constraints. Mine literally says my capacity is part-time and that this is a design requirement, not a temporary state. That one line changes every single answer I get back.</p><p><strong>2. What the business is.</strong> Not my title. The actual spine of the work, in a sentence I’d say out loud. Mine is: extract how someone really works and sounds, build the system around that reality, then teach them to run it themselves.</p><p><strong>3. How I talk about it.</strong> Two versions of my pitch, because I need a different one for a room of moms than for a room of founders. Written in my words, not polished marketing words.</p><p><strong>4. What I sell and what it costs.</strong> Every offer, every price, and honestly whether each one is real or still an experiment. Mine has a column that just says live, experimental, or retire this one.</p><p><strong>5. Who I say yes to, and who I don’t.</strong> The clients that energize me and the specific signals that mean decline even when money is on the table. Writing that down changed how fast I say no.</p><p><strong>6. Capacity and money.</strong> What a good month looks like, what I’m aiming at, and what’s actually true right now instead of what I wish were true.</p><p><strong>7. How to work with me.</strong> My rules. Don’t propose rebuilding things that work. Don’t delete anything without showing me the list first. Small and reversible over big and impressive.</p><h3>The section I left blank</h3><p>Mine has a section called needs Eleah. It holds the questions I have not answered yet, like what I actually value, and what I want this business to make possible in my life.</p><p>I leave those blank and label them open, rather than writing something that sounds finished. Every agent I run reads this file. I would rather they know a question is still unanswered than act on an answer I invented to fill the space.</p><p>Also, keep it separate from your day-to-day status. I have one file for who I am, which barely changes, and a second for what’s true this week, which changes constantly. When I mixed them, both went stale.</p><h3>Easiest way to write it: don’t.</h3><p>Everyone hears “write a document about yourself” and their whole system kind of goes tight, because now it’s homework and there’s a right way to do it.</p><p>So talk out loud to your AI about your life and work for twenty minutes and have it draft the thing, then correct what it got wrong. Correcting is much easier than starting from a blank page, and the corrections are where it actually gets good.</p><p>You do not have to start from a blank page, and you do not have to copy my structure. Paste the block below into a fresh conversation with whatever AI you use, and let it pull the answers out of you.</p><blockquote>You are going to help me write a master prompt: a document about me that every future conversation with you will inherit. Interview me for about twenty minutes, one question at a time, and do not move on until my answer is specific. Push back when I am vague. Cover these: who I am, including the real constraints on my time and energy; what my business actually does, in a sentence I would say out loud; how I describe it to two different audiences; what I sell, what it costs, and whether each offer is real or still an experiment; who I say yes to, and the signals that mean I should decline; my capacity, and what a good month looks like; and how I want you to work with me, my rules. When we are done, draft the document in my words, not in polished marketing language. If I could not answer something clearly, do not fill it in. Mark it as an open question so we both know it is unanswered.</blockquote><p>That last instruction is the one that matters most. An AI will happily invent a plausible answer for a question you have not resolved, and then every conversation after that inherits the invention.</p><h3>Who I learned this from.</h3><p><strong>Dan Martell</strong> for the framework. He’s a little bit the classic AI tech guy, but he put language to something I’d felt and hadn’t named: the shift from AI answering you to AI working with you. Chat, connectors, agents. He’s also where the term master prompt comes from for me.</p><p><strong>Brock Mesarich (AI for Non Techies)</strong> for keeping it simple. No jargon, no assumed background. His Claude Cowork walkthrough covers the exact tool most people are asking about right now.</p><p><strong>Nate Herk (AI Automation)</strong> for when you want to build. More technical, agent building and workflow tools. Useful once you’re past “what even is this.”</p><p>I wrote the longer version with the specific videos and why each one earns the click over here: <a href="https://leala.co/articles/who-i-learn-ai-from">Who I learn AI from</a>.</p><p>And I talk through the three levels in my own voice in <a href="https://open.spotify.com/episode/6i86utcwxfHgt4BYoqzxjE">episode two of my podcast</a>, if you’d rather listen than read. I started it because I think more women should be talking about AI out loud and imperfectly.</p><p><strong>One hour. Go talk it into existence.</strong></p><p><em>Originally published at </em><a href="https://leala.co/articles/master-prompt-whats-in-mine.html"><em>https://leala.co</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1b1782a18eab" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Who I Learn AI From]]></title>
            <link>https://medium.com/@eleahminer/who-i-actually-learn-ai-from-d12a5bccb521?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/d12a5bccb521</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[women-in-tech]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Fri, 21 Aug 2026 18:40:29 GMT</pubDate>
            <atom:updated>2026-08-24T18:07:49.879Z</atom:updated>
            <content:encoded><![CDATA[<h3>The guy who gave me the framework.</h3><p>This is where I learned to think in three levels of AI: chat, connectors, and agents.</p><p>I think more women should be talking about AI, out loud and imperfectly. I started my own podcast about it a few weeks ago. <a href="https://open.spotify.com/episode/6i86utcwxfHgt4BYoqzxjE">Episode two</a> covers the exact three levels I’m about to describe, if you want the longer version in my own voice.</p><p>I don’t have anywhere near the bandwidth the people below have. They do this full time. I very much do not. But here’s who actually taught me the things I now say out loud.</p><p>Dan Martell is a little bit the classic AI tech guy. But his big picture thinking is worth the watch. He’s the one who put language to something I’d felt but hadn’t named. The shift in how you get AI to work with you, instead of just answering you. Chat, connectors, agents. That’s the whole framework, and it’s the clearest version of it I’ve heard anywhere.</p><p>One honest note. This was recorded back in March. A lot has changed since then. The framework still holds even where the specific tools have moved on.</p><p>Dan Martell · The AI Cheat Codes Every Founder Needs in 2026</p><p><a href="https://youtu.be/_24HzGNv-3A">Watch the video →</a></p><h3>The one who keeps it simple.</h3><p>AI for Non Techies does exactly what the name says. No jargon, no assumed background.</p><p>Brock Mesarich runs a channel called AI for Non Techies, built for people who don’t come from a technical background and don’t want to pretend they do.</p><p>The video I’d point you to walks through every Claude Cowork concept end to end. It happens to be the exact tool most of us on that call were asking about.</p><p>Brock Mesarich · AI for Non Techies</p><p><a href="https://www.youtube.com/watch?v=NDHWUhGzKg0">Watch the video →</a></p><h3>If you want to go deeper.</h3><p>More technical. Workflow and automation focused. Good once you’re ready to build.</p><p>Nate Herk covers AI automation, agent building, and workflow tools like n8n. It’s a step further into the technical weeds than Brock’s channel.</p><p>Useful once you’re past the what even is this stage. Ready to start building things yourself.</p><p>Nate Herk · AI Automation</p><p><a href="https://www.youtube.com/@nateherk">Visit the channel →</a></p><p><em>Originally published at </em><a href="https://leala.co/articles/who-i-learn-ai-from.html"><em>https://leala.co</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d12a5bccb521" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Where to start with AI.]]></title>
            <link>https://medium.com/@eleahminer/where-to-start-with-ai-4bb41f886047?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/4bb41f886047</guid>
            <category><![CDATA[productivity]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Thu, 13 Aug 2026 19:07:46 GMT</pubDate>
            <atom:updated>2026-08-24T17:43:36.841Z</atom:updated>
            <content:encoded><![CDATA[<p>Yesterday I was somewhere in Nevada, driving my son home from a summer in California, when a community I’m part of asked me to hop on their call and talk about AI.</p><p>The question that kept coming up was the same one I hear everywhere now. Where do I even start? There are a thousand tools. Everyone’s overwhelmed. Half the people asking have already tried ChatGPT, gotten something generic back, and quietly decided AI isn’t for them.</p><p>I want to tell you what I told them, because I think the answer is smaller and less technical than anyone expects.</p><p>You don’t start with a tool. You start with a document.</p><h3>The document that changes everything</h3><p>Before I explain, some context. I’m not an engineer who fell into wellness or a coach who fell into tech. I’m both at once, somatic guide and software engineer, and for the past few years AI has been the thing holding my life together as a busy mom building a business without a team. My AI edits my podcast. It organizes my files. It helped me write this on a road trip.</p><p>None of that started with an app. It started the day I wrote down who I am.</p><p>Dan Martell calls it a master prompt. It’s one document that tells AI everything it needs to know about you: what you do, who you serve, how you like to work, what you’re building, and what your real constraints are. Mine says “busy mom” in it, and I’m not kidding when I say those two words change every answer I get.</p><p>Here’s why this matters. When AI gives you generic advice, it’s usually not because the tool is dumb. And it’s not because you’re dumb. It’s because it knows nothing about you, and you keep introducing yourself from scratch, ten times a day, forever. So it treats you like the average of everyone. Of course the answers feel hollow.</p><p>A master prompt ends that. Every conversation starts already knowing you.</p><p>And you don’t even have to write it. Talk it. Open a conversation with your AI and ramble about your life and work for twenty minutes, the way you’d catch up an old friend. Then ask it to draft your master prompt from what you said, and correct what it got wrong. The correcting is where it gets good. One hour, total.</p><p>That hour does more than any subscription you could buy. I really mean that.</p><p>And if you’ve already been using AI for a while, there’s a faster on-ramp. You’ve technically been writing your master prompt for months, one conversation at a time. So export your history from ChatGPT or Gemini or whatever you’ve been living in, and hand it over. Point it at your LinkedIn. Give it your resume. Then ask it to write the master prompt from everything it now knows about you. You’re not starting from scratch. You’re harvesting.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*07mEpFmlo9Kcno4OZvRr2Q.png" /></figure><h3>Where your thoughts go to be useful</h3><p>Step two is what I call my open brain, and it’s the thing people on the call wanted to understand.</p><p>A master prompt tells AI who you are. A second brain gives it access to what you know and think. Without one, every insight you have in the shower or the car just evaporates. With one, your thinking accumulates somewhere your AI can search.</p><p>Mine is a database my AI reads and writes for me. I talk. Voice notes while driving, thoughts after a client call, ideas mid-walk. The system captures each one and files it: what kind of thought it is, what topic it touches, who it involves. I pulled the numbers this morning and there are 778 thoughts in there since March, organized across topics like AI, co-parenting, coaching, scheduling. I organized almost none of it by hand.</p><p>When I ask my AI to help me with something, it can search all of that. My own thinking, in my own words, from months back. That’s the moment AI stops feeling like a clever stranger and starts feeling like it knows you. Because now it does.</p><p>The other path is Obsidian, which a few people on the call had heard of. Your notes live as plain files in folders you own, on your computer, and as of this year Claude connects to an Obsidian vault really well. When I built my open brain, that integration barely existed, so I made my own out of a database. Today both paths work, and under the hood they’re the same idea: a memory your AI can search.</p><p>So the real question isn’t which tool. It’s which one is shaped like you.</p><p>If you think by talking and want the filing done for you, build the open brain style. If you think by writing and love linking your own notes together, Obsidian will feel like home, and AI rides along.</p><p>There’s no wrong answer here. The wrong answer is having neither, and re-explaining your whole life to a goldfish every morning.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*U9VMg1P38JpaXfK0Kq9bfA.png" /></figure><h3>What this unlocks</h3><p>Once AI knows who you are and can reach what you think, the impressive stuff stops being impressive. It just becomes how things work.</p><p>Two weeks ago I launched a podcast. My job was to talk into a recorder. My AI cleaned the audio, removed my mouth clicks, auditioned intro music with me until something in my body said yes, mixed it, mastered it, and handed me a finished episode while I made lunch. A year ago that’s three professionals I couldn’t afford. This year it’s a Tuesday.</p><p>I don’t say that to be impressive. I say it because none of it required me to be technical that day. It required the foundation to exist. The system knew my voice rules, my brand, my podcast, my patterns, because I’d done step one and step two months ago.</p><p>And here’s what I’ve learned about building the workflows themselves. There are two ways, and both are right.</p><p>The intentional way: pick one problem, design the solution step by step, careful and deliberate. If that’s your temperament, beautiful.</p><p>The messy way, which is mine: do the thing once, badly, with your AI doing as much as possible. Let it be imperfect. Then look at what actually worked, correct the output until it’s what you wanted, and reverse-engineer that into a saved skill your AI runs the same way every time. Claude has skills for this, GPT has its own version. The mess is the blueprint.</p><p>The podcast was the messy way. The imperfection gives me a little bit of an ick. The sound isn’t where I want it. My energy in episode one isn’t where I want it. I published anyway, because the habit matters more than the polish, and because every single week the system gets a little better while costing me a little less. That’s the metric I care about now. Not perfection. What my limited time buys me, this week versus last week.</p><p>And I think this is what’s underneath so much AI anxiety right now. People are starting at step three. They’re trying to get magical outputs from a tool that doesn’t know them, getting slop back, and concluding the whole thing is hype or somehow beyond them. Some people call the exhaustion of that loop AI burnout. But you can’t burn out on a relationship that never actually started.</p><p>What I keep learning is that this was never a tech problem. It’s a self-trust problem. The tool is waiting for you to tell it who you are, and most of us have never sat down and said it plainly, even to ourselves. Which might be why that first hour feels bigger than it should.</p><h3>If you’re starting this week</h3><p>Keep it this small:</p><ol><li>One hour. Talk your master prompt into existence, let AI draft it, correct it, save it somewhere you can paste from.</li><li>Pick your lane. Talker: a simple capture system your AI files for you. Writer: an Obsidian vault connected to Claude.</li><li>Capture for two weeks before you judge anything. The magic isn’t in day one. It shows up when there’s enough of you in there to search.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bMZB82DDVY_J1QG_bgMaog.png" /></figure><p>That’s it. No course, no stack of subscriptions, no becoming a different kind of person.</p><p>I’m writing this from the passenger seat somewhere past Tahoe. My son is in the back asking questions about the mountains, school starts tomorrow, the car needs to go to the shop, and the week ahead feels like a lot. But somewhere in a database, a running record of my thinking is waiting for me, and the system I started with one document months ago will catch most of what this week throws at it.</p><p>Because here’s what I’m building toward, and I might as well say it out loud. A living, breathing diary of my life and work that turns itself into resources people can use. So that the way I get clients is by being myself, in public, consistently, while the system does the shaping and the shipping. I think that’s quietly the dream for most business owners: not to market harder, but to be findable as yourself. I’m not there yet. I’ve been too buried in client work to even post regularly, and I’m only now coming back to it, imperfect on purpose.</p><p>But the foundation makes it possible. It doesn’t replace you. It remembers you.</p><p>Start with the document.</p><p><em>It doesn’t replace you. It remembers you.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4bb41f886047" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[One Hour Over Coffee Changed How She Uses AI]]></title>
            <link>https://medium.com/@eleahminer/one-hour-over-coffee-changed-how-she-uses-ai-5341baf5c428?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/5341baf5c428</guid>
            <category><![CDATA[mindfulness]]></category>
            <category><![CDATA[mental-health]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[self-improvement]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Thu, 16 Apr 2026 13:01:01 GMT</pubDate>
            <atom:updated>2026-04-16T13:01:01.960Z</atom:updated>
            <content:encoded><![CDATA[<p><em>On overwhelm, self-trust, and the difference between learning a tool and building your own system.</em></p><p>So I had coffee with a friend of a friend yesterday. Let’s call her Jess. She sat down, took her boots off, and said something like, “I use AI a little, but I feel like I’m doing it wrong.”</p><p>And I get that. I hear some version of that sentence almost every week now. People see the posts. They see the dashboards. They hear about agents and automations and prompt engineering and they think, okay, I need to learn all of this or I’m going to be left behind. Especially solopreneurs. Especially the ones running everything themselves already. That feeling, that gap between where you are and where it seems like everyone else is, it lives in the body. It’s not just an information problem. It’s a nervous system problem.</p><p>Jess has a full life. She teaches a class at a university, she has an executive coaching client, she’s pitching herself for operations work at a law firm, and she’s nursing a baby who just started daycare. Yesterday was literally her first free morning in over a year.</p><p>She didn’t need a tutorial. She needed someone to listen.</p><h3>What she actually said</h3><p>Here’s what I’ve learned from doing this a few times now. When someone says “I want to learn AI,” they don’t actually mean that. What they mean is, “I have too much on my plate, and I heard this thing could help, but I don’t know where to start and honestly I’m a little scared it’s going to make me feel stupid.”</p><p>Jess had been using ChatGPT for her university lesson plans. And it worked, kind of. But she told me that the times she phoned it in, just took what it gave her, those lessons fell flat. The students could feel it. She could feel it. And so she’d learned this thing that I think a lot of people are learning right now, which is that AI is not a shortcut. It’s a much more active process than people expect.</p><p>But she didn’t know how to make it more active. She just knew the default wasn’t working.</p><p>And then with her coaching client, she’d tried to use Claude for session prep, but it kept giving her these generic frameworks. Basic stuff. Not her methodology, not the way she actually moves through a coaching session. So she’d kind of given up on that.</p><p>And the law firm thing, she hadn’t even started to think about how AI would fit into that yet.</p><p>So she’s sitting across from me with three different work lanes, each of which could use AI differently, and she’s overwhelmed. Not by the technology. By the distance between where she is and where she thinks she needs to be.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l5-whsNqdOqbI5PPx2er8Q.png" /></figure><h3>What I actually did</h3><p>I did what I always do. I listened. And then I talked to my team.</p><p>I have a system I’ve been building for about a year now. It’s a set of AI agents, each one trained for a different kind of work. One of them is really good at taking in a person’s context and mapping out where AI can help first. So while Jess and I were talking, I was also feeding what she was telling me into that system. Not word for word. Just the shape of what she was saying. Her work lanes. What was frustrating her. What she actually needed.</p><p>And within about fifteen minutes, my system came back with a plan. Not a generic “here’s how to use AI” plan. A specific one, for her, based on the things she’d just told me.</p><p>It said: start with the university class. It’s the lowest stakes, the most immediate, and it’ll build her confidence fast. Create a project in Claude with her syllabus, her teaching philosophy, her grading approach baked in, so the AI stops guessing and starts thinking like her.</p><p>That’s the whole trick, by the way. Most people are using AI with no context. They open a chat window and ask it to help them and they’re surprised when the output is generic. But if you give it your voice, your frameworks, your “here’s how I actually do this,” suddenly it stops being a tool and becomes an extension of how you already think.</p><p>So that’s what we set up. In about twenty minutes, Jess had a project configured for her class with enough context that when she asks for help with a lesson plan, it comes back sounding like her. Not like a robot. Not like a textbook. Like her.</p><p>She looked at me and said, “I want to learn how to do this myself.” Which is exactly the right answer. I can build things for people, but the ones who actually stick with it are the ones who want to understand it enough to build their own.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7ieoa_KnPoGMB4TWy8FHuA.png" /></figure><h3>The pattern</h3><p>Here’s the thing I keep noticing. The initial feeling is always the same. It’s overwhelm. It’s “I should already know how to do this.” It’s a low hum of inadequacy that sits right in the chest.</p><p>And the fix is also always the same. You don’t start with the technology. You start with the person. What are you actually trying to do? Not what do you think you should be doing with AI. What’s actually on your plate that’s too much?</p><p>I’ve done this with a nonprofit director who was drowning in grant writing and couldn’t figure out why Claude’s output felt so flat. Turns out she’d never told it how she writes. We spent our first session just defining her voice, and suddenly the tool started working.</p><p>I’ve done it with an author who has eighty hours of interview recordings and no way to organize them. She didn’t need an AI tutorial. She needed someone to listen to what was overwhelming her and then show her what was possible.</p><p>And now Jess. Who walked in overwhelmed and walked out with one clear starting point and the confidence to tinker with it.</p><p>One hour. Coffee. No slides, no demo, no pitch deck.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/675/1*Pn9m6aKaVskkhSh0z1bacQ.png" /></figure><h3>Why this matters to me</h3><p>I keep thinking about the gap. Not the skills gap that people write about on LinkedIn. The trust gap.</p><p>Most people don’t trust themselves with new technology. And honestly, a lot of the AI world is set up to make that worse. Every week there’s a new tool, a new framework, a new thing you’re supposed to be using. People are calling it AI burnout now. But I don’t think burnout is the right word. Burnout implies you’ve been doing too much. This is more like freeze. You haven’t even started, and your body is already exhausted by the distance. And if your nervous system is already at capacity, adding more information just makes that worse.</p><p>So I’ve been doing something really simple. I sit down with people and I listen. I figure out where they actually are, not where the internet says they should be. And then I build them one thing. One small thing that works immediately and that they can expand on their own.</p><p>It’s not scalable advice. It’s not a course. It’s just coffee and a question: what’s actually on your plate right now?</p><p>And then we build from there.</p><blockquote>You can’t solve a self-trust problem with a YouTube tutorial. You solve it by showing someone, in real time, that the thing they’re scared of is actually something they can do.</blockquote><p>I think the reason this keeps working is because it respects where someone’s body is. Jess didn’t need more information. She needed to feel like she could start. That’s a self-trust problem. And you can’t solve a self-trust problem with a YouTube tutorial.</p><p>You solve it by showing someone, in real time, that the thing they’re scared of is actually something they can do. And then you let them do it.</p><h3>What Jess texted me after</h3><p>She texted me a few hours later. She’d been tinkering. She’d set up a second project for her coaching practice. She was already playing with how to get Claude to think more like her when prepping for client sessions. She said something like, “I can’t believe how much time I wasted being scared of this.”</p><p>And I thought, you didn’t waste time. You just didn’t have someone sit down with you and make it small enough to start.</p><p>That’s what I want to keep doing. I’ve started calling it an AI Clarity Session, which is a fancy name for what is really just coffee and a question. One hour. You tell me what’s on your plate. I listen. We build one thing together before you leave.</p><p>If that sounds like where you are right now, I have a few spots open. <a href="https://calendar.app.google/DSmCTSVA64oAf7dy8">Let’s have coffee.</a></p><p>Start where you are.</p><p><em>This is the fifth in a series about building with AI as a non-technical founder. The </em><a href="https://medium.com/@eleahminer/what-my-ai-team-actually-does-ecdad70a2375"><em>first</em></a><em>, “What My AI Team Actually Does,” covers the architecture. The </em><a href="https://medium.com/@eleahminer"><em>fourth</em></a><em>, “I Built an AI Team. Then I Went Door to Door,” covers the depth-trap. This one covers what happens when you sit down with someone and make AI small enough to start.</em></p><p><em>I’m building </em><a href="https://leala.co/"><em>LEALA</em></a><em>, a somatic healing and inner voice community. Everything I’m learning about AI, nervous system regulation, and creating from alignment gets shared in real time.</em></p><p><em>If something in this piece landed, </em><a href="https://calendar.app.google/DSmCTSVA64oAf7dy8"><em>let’s talk</em></a><em>.</em></p><p><a href="https://www.linkedin.com/in/eleah-miner/"><em>LinkedIn</em></a><em> · </em><a href="https://www.instagram.com/leala.io/"><em>Instagram</em></a><em> · </em><a href="https://www.youtube.com/@leala.io"><em>YouTube</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5341baf5c428" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Built an AI Team. Then I Went Door to Door.]]></title>
            <link>https://medium.com/@eleahminer/i-built-an-ai-team-then-i-went-door-to-door-47191535a7f9?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/47191535a7f9</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[mental-health]]></category>
            <category><![CDATA[mindfulness]]></category>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Tue, 14 Apr 2026 14:01:03 GMT</pubDate>
            <atom:updated>2026-04-14T14:01:03.774Z</atom:updated>
            <content:encoded><![CDATA[<p><em>On self-trust, nervous system regulation, and the depth-trap nobody is naming yet.</em></p><p>I feel a little nervous writing this, because the part of me that wants to tell you the truth about AI and my heart behind all of it doesn’t really seem like what people want to hear. Everyone wants the system. The dashboard. The before-and-after. And I have those. I built those. I can show you a screen with eight agents running at the same time.</p><p>But that’s not the part I want to write about today.</p><p>The part I want to write about is what happened last Tuesday, when I printed a one-page resume on my own printer, folded it in half, put it in my bag, and walked into a coffee shop in Boulder to ask for a job.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*veqLRW40EcnGa50CXX3ljg.png" /></figure><h3>The depth-trap</h3><p>Here’s the thing nobody told me about building with AI. The tool gets better every day. Which means I can create at a much higher level every day. Which means I can share more deeply every day. Which means, if I am not careful, I can get stuck in a loop where I keep rewriting the same thing because I can go one layer deeper than I could yesterday.</p><p>That’s the trap. Not AI burnout, not the kind people write about on LinkedIn. Something quieter.</p><p>I’m calling it the depth-trap, because I haven’t heard anyone else name it yet and I think it is about to become the default condition for anyone who uses AI seriously. Harvard Business Review calls the broader version of this “AI brain fry.” But that label puts it in your head. Mine lives in the body.</p><p>The depth-trap works like this. You write something. It’s good. Then you notice the tool is better today than it was yesterday, so you rewrite it, and it’s a little deeper. Then you notice that new depth revealed another layer, so you rewrite it again. And again. At some point you look up and three days have passed and the thing you were going to post is still sitting in a draft.</p><p>The rewriting isn’t the problem. I love the rewriting. The problem is that my body loses the floor. There’s no stopping point the system can give me. Only my nervous system can tell me when I’m done. And my nervous system can’t tell me anything if I never stop long enough to listen. That’s a self-trust problem. Not a productivity problem.</p><p>So I stopped.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UiiLwBLg6IXxIKqIhlXXAQ.png" /></figure><h3>The three things I refuse to do with AI</h3><p>I want to name these out loud because if I don’t say them, I will drift. The tool is too good. The drift is too easy.</p><p>One. I refuse to use AI to just go outward. To just pump content, content, content without going inside myself first. I won’t do it. If I can feel that I’m not sure why I’m making the thing, I don’t make the thing. That’s the rule.</p><p>Two. I refuse to use AI to promise things that aren’t true. I had hopes, real hopes, that my agent team could be fully autonomous. It can’t be. Not yet. Not at the level I would need to hand a client a system and say “this runs itself.” So I’m not going to sell that to anybody. I’m testing all of this on myself first. I’m the experiment. Everyone I work with later gets a version of what I have already learned to hold with my own body.</p><p>Three. I refuse to make anything I wouldn’t want to watch. I get the ick watching most of what’s out there right now. You can feel when a person is posting with care. You can feel when they aren’t. I don’t want to be on the wrong side of that. I would rather post less and mean it than post a lot and drift.</p><p>That’s the whole list. Three lines. I’m going to read them every morning.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IQmaxXSH_O1h69DwC8NsVA.png" /></figure><h3>Then I went door to door</h3><p>So here’s what I actually did last week, while most people in my position were posting hot takes about the next AI tool.</p><p>I made a one-page resume. I called it my Ultimate Generalist one-sheet. It had two sections. “On the Floor” for the front-of-house work, the hospitality, the warmth. “Behind the Scenes” for everything else. Web design, content, ops, tech support. I listed what I can actually do, and I wrote at the top that I was looking for fifteen to twenty hours a week in a grounded, local place.</p><p>Then I printed it. I folded it in half. I put it in my bag.</p><p>And I walked into businesses in Boulder and handed it to people.</p><p>My chest was nervous the whole time. I’m not going to pretend otherwise. I have a tech background. I have a portfolio. I have an AI system running at home that most people couldn’t build in a year. And I was standing in a coffee shop asking a stranger if they needed a few shifts covered.</p><blockquote>The next day, one of them offered me a position.</blockquote><h3>Why I needed the door</h3><p>People are going to ask me why. I want to try to answer from the part of me that actually knows.</p><p>My nervous system needed stability.</p><p>That’s the whole sentence. Everything else is explanation.</p><p>I have been inside the AI rollercoaster for months. Building agents. Testing systems. Holding the edge of what’s possible and what doesn’t work yet. It is the most exciting work I have ever done. And it is also the work that asks the most from my body, because nothing is ever finished and the tool keeps getting better and the depth-trap keeps opening underneath me.</p><p>My body wanted a floor. Not a metaphorical floor. An actual floor. A place I could walk to. A place where someone hands me a list of things to do and I do them and I go home. A place where I can bring the warm, front-of-house part of me that doesn’t live inside any of my dashboards. A place my nervous system recognizes as ground.</p><p>This is the part the AI productivity crowd isn’t going to talk about. Going slower is not an AI content strategy. Going slower is what your body will force you to do if you don’t do it on purpose first. I chose to do it on purpose.</p><h3>The skill nobody is naming</h3><p>Everyone is talking about the AI skills. Prompting. Building. Automating. Scaling. These are real skills, and I use all of them every day.</p><p>But there is another skill, and it is the one that’s actually going to matter most in the next five years, and almost nobody is naming it.</p><p>It is the skill of knowing when to stop.</p><p>Knowing when your nervous system is asking you to walk outside and touch a door. Knowing when the output is already good enough and another pass of depth would cost you your floor. Knowing when the right move is to close the laptop and take a piece of printed paper to a physical building full of physical people.</p><p>I don’t think this skill is soft. I think this is the hardest skill of the whole AI era. Because the tool will never tell you to stop. The tool is infinite. Only your body knows where the edge is. And if you have outsourced your sense of “enough” to a system, you will never feel it arrive.</p><p>Self-trust is the muscle that knows when to stop. That is what I am practicing. That is what the one-sheet was.</p><h3>Four questions I am sitting with</h3><p>I am not going to tell you to do what I did. I don’t believe in recipes. I am going to ask you four questions, because questions move me more than answers do.</p><p><em>Where in your work are you going deeper because you can, rather than because you need to?</em></p><p><em>What is one thing you have been rewriting this week that your body already knows is finished?</em></p><p><em>If your nervous system could ask you for one piece of stability right now, what would it be?</em></p><p><em>What is the physical version of the door you need to walk through?</em></p><p>You don’t have to answer these here. You don’t have to tell anyone. But I would sit with them. I am sitting with them too. I am still in the middle of this.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lRC3XYcEZG4PuHCHObFX7w.png" /></figure><p>I am surviving. I am showing up every day. I am facing my fears. I am doing the things nobody else is doing, because I want to be excellent at what I do, and excellence is not the same thing as constant output. Excellence is the muscle that knows when to stop.</p><p>I built an AI team. Then I went door to door. Both are true. Both are the work.</p><p><em>This is the fourth in a series about building with AI as a non-technical founder. The </em><a href="https://medium.com/@eleahminer/what-my-ai-team-actually-does-ecdad70a2375"><em>first</em></a><em>, “What My AI Team Actually Does,” covers the architecture. The </em><a href="https://medium.com/@eleahminer"><em>second</em></a><em>, “I Didn’t Use AI to Be More Productive,” covers the why. The </em><a href="https://medium.com/@eleahminer"><em>third</em></a><em>, “I Used AI to Hear My Own Voice,” covers what it feels like when the system finally lets you land on what you actually want to say. This one covers the part where I had to stop, fold a piece of paper in half, and walk into a building.</em></p><p><em>I’m building </em><a href="https://leala.co/"><em>LEALA</em></a><em>, a somatic healing and inner voice community. Everything I’m learning about AI, nervous system regulation, and creating from alignment gets shared in real time.</em></p><p><a href="https://www.linkedin.com/in/eleah-miner/"><em>LinkedIn</em></a><em> · </em><a href="https://www.instagram.com/leala.io/"><em>Instagram</em></a><em> · </em><a href="https://www.youtube.com/@leala.io"><em>YouTube</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=47191535a7f9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I’m Applying Atomic Habits in the Age of AI]]></title>
            <link>https://medium.com/@eleahminer/how-im-applying-atomic-habits-in-the-age-of-ai-e03966ddea51?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/e03966ddea51</guid>
            <category><![CDATA[systems-thinking]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[claude]]></category>
            <category><![CDATA[james-clear]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Wed, 01 Apr 2026 14:24:40 GMT</pubDate>
            <atom:updated>2026-04-01T14:24:40.825Z</atom:updated>
            <content:encoded><![CDATA[<p><em>James Clear’s framework clicked in a way it never had before. Here’s what I’m building with it.</em></p><p>I picked up Atomic Habits yesterday, wanting to be inspired to create some new habits to support my goals and systems. And I kept having this moment of, wait. I’ve been living this, just in a completely different language.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2hlRiZte6U7jTlzpldItIQ.png" /></figure><p>James Clear breaks behavior change into three layers. The first layer is outcomes, which is what you get. The second is processes, which is what you do. And the third, the deepest one, is identity, which is what you believe.</p><blockquote><em>Your worldview, your self-image, your judgments about yourself and others, most of the beliefs, assumptions, and biases that you hold are associated with this level. -James Clear</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3IYAiRX_CrOHhd7qB-DVcQ.png" /></figure><p>And then he makes this distinction that kind of rewired how I think about goals: <em>“Goals are about the results you want to achieve. Systems are about the processes that lead to those results.”</em></p><p>The way he explains it, when you only focus on goals, you’re basically setting yourself up for a pass/fail test. You either hit the goal or you didn’t. And every day you haven’t hit it, there’s this quiet signal running in the background telling you that you’re not there yet. That you’re failing. Even if you’re making progress, the goal itself creates this gap between where you are and where you think you should be.</p><p>Systems flip that. Instead of fixating on the outcome, you build the environment and the processes that make the outcome kind of inevitable. Clear uses this example over and over: instead of setting a goal to be healthy, just ask yourself, what would a healthy person do right now? And then do that.</p><p>That’s the part that got me. Because I’ve been using a framework from Bob Proctor for years to set my goals, and it suddenly clicked in a new way.</p><h3>PROCTOR’S A, B, AND C GOALS</h3><p>If you’re not familiar, Bob Proctor talks about three types of goals. An A goal is something you already know how to do. You’ve done it before, you know the steps. A B goal is something you think you can probably figure out. It’s a stretch, but it’s within reach. And a C goal is the one that actually scares you. It’s the thing you really want, but you have no idea how to get there. You’ve never done it. It might feel like a fantasy.</p><p>Proctor says the C goal is the only one worth setting. Because it requires you to become someone you haven’t been yet.</p><p>And I love that framework. I use it. I have my A, B, and C goals mapped out. But here’s where it always got stuck for me. I’d set the C goal, feel inspired for a day, and then wake up the next morning staring at this massive gap between where I am and where that goal lives. And instead of feeling motivated, I’d feel overwhelmed. Like, okay great, I want that. Now what?</p><p>Clear gave me the now what.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VRMRpVoUwB3dVEunTrZnrw.png" /></figure><h3>SYSTEMS AS ENVIRONMENT</h3><blockquote>“You do not rise to the level of your goals. You fall to the level of your systems.”</blockquote><p>When I read that, I thought about every C goal I’ve set that didn’t go anywhere. And it wasn’t because the goal was wrong. It was because I didn’t build anything around it. I just had the goal sitting there, looking at me, and I was looking back at it like, so… how?</p><p>What Clear is really saying is: stop trying to close the gap through willpower. Build the environment that makes the gap irrelevant. Create systems that hold you at the level of the person who already has that thing.</p><p>And here’s where it clicks for me in a way it wouldn’t have even a year ago.</p><h3>HOW I’M BUILDING SYSTEMS NOW</h3><p>I’ve been building what I think of as my AI team. Basically, I’ve set up a system where different parts of my business are handled by AI that already knows how I work. One piece handles client outreach. One handles my content. One keeps my voice consistent across everything I publish. One handles the life logistics, the medical paperwork, the financial tracking, the co-parenting coordination.</p><p>Each piece of that system has my context, my preferences, and my standards built into it. So instead of waking up every morning and deciding from scratch how to tackle my C goal, the environment is already set up. It’s already holding the shape of what I’m building toward.</p><p>And here’s the part that really maps to Clear: every single time I use a piece of that system, I make it 1% better.</p><p>I look at what came out. I notice what’s off. I tweak it. And the next time it runs, the output is a little sharper. The templates are a little cleaner. The voice is a little more mine. It catches one more thing I hadn’t thought of before.</p><blockquote>“Every action you take is a vote for the type of person you wish to become.”</blockquote><p>That’s Clear. And the way I read it now is: every time I refine part of my system, I’m casting a vote for the version of me who already has what the C goal describes. Not through discipline. Through architecture. Through curiosity about what’s 1% off and what could be better.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Lyv_jmnlS6ufgwZ1FmREoQ.png" /></figure><blockquote>It’s building the environment of that person and then living inside it while you grow into it.</blockquote><h3>THE PIECE THAT GOES DEEPER</h3><p>So Clear gets to identity. He says the deepest layer of change is about what you believe. And I agree.</p><p>But I think there’s something underneath that he doesn’t quite name.</p><p>You can build the most elegant system in the world. You can have the agents, the workflows, the 1% improvements compounding. And if your body, not your mind but your actual nervous system, doesn’t believe you’re allowed to have the thing you’re building toward, you will find a way to not use the system. You’ll procrastinate on the one task that actually moves the needle. You’ll self-sabotage. I’ve done this. More than once.</p><p>There’s a concept in somatic work called an expander. An expander is a person who has the thing you want, and whose existence proves to your nervous system that it’s possible. Not just intellectually. Your body has to see it and register, oh. That person did it. That’s real. Someone like me can have that.</p><p>Because here’s what actually happens when you set a C goal. Your conscious mind goes, yes I want that. And your nervous system goes, we’ve never had that before. That’s not safe. The body doesn’t care about your vision board. It cares about pattern recognition. And if the pattern it knows is scarcity, or things falling apart right when they start working, then no system on earth will override that without the internal work.</p><p>So you need both. The architecture and the integration. You need the system that holds your 1% better environment. And you need the expander, the real human evidence, that rewires what your body believes is available to you.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PB-tTkYhCS5OMAK-IGVBSA.png" /></figure><h3>BUILDING IN REAL TIME</h3><p>I’m not writing this from the other side. I haven’t figured it all out. I’m building this in real time. I use other founders as my expanders. People who’ve made this kind of bridge work, the tech and the body and the business all in one practice. I watch how they move and I let my nervous system take notes.</p><p>And I’m documenting my own architecture so that I can be that expander for someone else. Not because I’m ahead. But because I’m willing to show the build.</p><p>That’s what this series is about. The actual architecture of someone applying frameworks like Clear’s and Proctor’s with AI, while also doing the somatic work that makes the identity shift real. Not just cognitive. Felt.</p><blockquote>“The purpose of setting goals is to win the game. The purpose of building systems is to continue playing the game.”</blockquote><p>I’m just playing it with tools he didn’t have yet. And with a body that’s learning to trust the game.</p><p><em>This is the third in a series about building with AI as a non-technical founder. The first, </em><a href="https://medium.com/@eleahminer/what-my-ai-team-actually-does-ecdad70a2375"><em>“What My AI Team Actually Does,”</em></a><em> is a behind-the-scenes look at the agent architecture. The second, “I Didn’t Use AI to Be More Productive. I Used It to Feel Safe Enough to Create,” covers the nervous system side. This one covers what happens when you combine James Clear’s systems thinking with AI and somatic integration.</em></p><p><em>I’m building </em><a href="https://leala.co/"><em>LEALA</em></a><em>, a somatic healing and inner voice community. Everything I’m learning about AI, nervous system regulation, and creating from alignment gets shared in real time.</em></p><p><a href="https://www.linkedin.com/in/eleah-miner-844969b8"><em>LinkedIn</em></a><em> · </em><a href="https://www.instagram.com/leala.io"><em>Instagram</em></a><em> · </em><a href="https://www.youtube.com/@leala-expansion"><em>YouTube</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e03966ddea51" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Didn’t Use AI to Be More Productive. I Used It to Feel Safe Enough to Create.]]></title>
            <link>https://medium.com/@eleahminer/i-didnt-use-ai-to-be-more-productive-i-used-it-to-feel-safe-enough-to-create-09bd865fee6e?source=rss-c65edfb60b19------2</link>
            <guid isPermaLink="false">https://medium.com/p/09bd865fee6e</guid>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[nervous-system]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Eleahminer]]></dc:creator>
            <pubDate>Wed, 01 Apr 2026 00:45:09 GMT</pubDate>
            <atom:updated>2026-04-01T00:45:09.916Z</atom:updated>
            <content:encoded><![CDATA[<p><em>How I built a nervous-system-first AI architecture, and why the inner voice had to come before the business plan</em></p><p>So I keep getting this question from friends. “What exactly is this AI thing you’re building?”</p><p>And I never know how to answer it. Because it’s not one thing. It’s more like I finally have a team. And they’re not human. And somehow that’s made my work more human than it’s ever been.</p><p>I should probably back up.</p><h3>I was drowning in friction</h3><p>I was in a season where everything shifted at once. Career. Relationship. Identity. The scaffolding I’d built my adult life on just wasn’t there anymore. And I was trying to figure out what to build next while also, you know, being a mom and keeping the lights on and not losing my mind.</p><p>And the friction was just everywhere. Not even the big emotional stuff. The logistical stuff. Medical appointments that take seven phone calls. Co-parenting schedules that change on Tuesdays. Content I wanted to create but couldn’t get to because I was buried in admin. Ideas I’d have at 6:30 in the morning that would just be gone by 8:00 because there was nothing to catch them.</p><p>When your nervous system is running that constant low hum of survival, you don’t just lose productivity. You lose access to yourself. The inner voice that used to help you make decisions gets buried under all the noise. And then you feel like a fraud, because the thing you most want to teach people about is the exact thing you can’t reach.</p><p>That’s where I was.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HH8pawcz-YSbPoIRFJ1u9A.png" /></figure><h3>I didn’t start with the tech</h3><p>Before I opened a single AI tool, I did something that probably doesn’t belong in a tech article. I sat with my nervous system.</p><p>I have a background in somatic work. Body-based practices that help people feel safe in their own skin. I’ve guided dozens of sessions helping people drop in and notice what’s actually happening in their bodies. I know, in my bones, that the body is where the real knowing lives.</p><p>So instead of starting with a business plan or a spreadsheet, I started with a Pinterest board.</p><p>I know how that sounds. But here’s what happened. I closed my eyes. I dropped into that place I go when I’m calming my nervous system. That quiet, textured space where everything slows down. And I started collecting images that matched the feeling. Warm stone. Soft light through tall windows. The grain of handmade paper. Sage and clay and cream. Nothing corporate. Nothing that vibrated at the frequency of hustle.</p><p>That Pinterest board became my design system. Not because I’m a designer. Because I know what regulation feels like, and I wanted my work to feel like that too.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Pht0E2kjJPd9hzPr-ZvpDA.png" /></figure><h3>Then I taught the AI to speak from that place</h3><p>This is the part that feels sacred to me. And I don’t use that word casually.</p><p>Prompt engineering is really an entire skill on its own. And for me it’s not a technical exercise. It’s an act of translation. You take what lives in the body, that felt sense of “this is right” and “this isn’t me,” and you learn to say it precisely enough that a machine can reflect it back.</p><p>I spent weeks training my AI system on my voice. Not just what I say, but how I say it. The way I circle back to an idea three times before I land on it. The small words I use when I’m being honest. The places I pause. The difference between my written voice and my spoken voice. And why the spoken one is almost always closer to the truth.</p><p>I built a voice guide. A visual mood library. A set of corrections that are non-negotiable: don’t smooth my edges. Don’t add transitions I wouldn’t say out loud. Let me circle. Let fire and softness exist in the same sentence. I documented how I actually think, how I process, how I show up when I’m being myself.</p><p>And then something happened that I wasn’t expecting.</p><p>The AI started giving me back language that sounded like me. Not a polished, corporate version of me. The actual me. The one who thinks out loud in the car and says “I don’t know” in the middle of a paragraph and uses the word “sacred” when talking about technology because she means it.</p><blockquote>I needed to remove the friction of survival so my inner voice could actually be heard. Not so I could scale a content operation. So I could hear myself think.</blockquote><h3>The whole system became a life raft</h3><p>I didn’t build this because my life was calm and I had time to optimize. I built it because I was drowning in friction.</p><p>When your nervous system is in survival mode, every task that requires executive function feels impossible. Responding to an email, scheduling a post, following up with someone you care about. Each one needs you to hold fifteen things in your head while your body is screaming that there’s a threat somewhere. And the gap between what you want to create and what you can actually execute gets so wide it starts to feel like evidence that you’re failing.</p><p>The AI agents I built aren’t productivity tools. They’re friction removers. They exist so that when I wake up at 6:30 with an idea that could change the direction of my week, there’s a system waiting to catch it. They exist so that the twelve-minute voice note I record, the one with the content idea tangled up with the co-parenting schedule tangled up with the client insight, gets untangled and routed to the right place without me sitting at a desk for three hours.</p><p>They exist so I can stay in my body and still have a functioning business.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xMRhtbVT65egl2PDRULyYw.png" /></figure><h3>What this actually looks like</h3><p>I wake up. I talk into my phone. That recording gets transcribed and processed and distributed to my AI team while I make coffee.</p><p>By the time I sit down, action items are sorted. Content drafts are waiting. My schedule is organized. The insights I had while walking, the ones that used to just vanish, are captured in a shared memory system that every agent can access.</p><p>But here’s what matters. The content that comes out doesn’t sound like a machine wrote it. It sounds like me. Because the whole system was designed around my voice, my nervous system, and my values. Not around some template I downloaded. Around the actual way I think and feel and process the world.</p><p>When something comes through that doesn’t feel right, and I can always tell, I don’t post it. My inner voice is still the final editor.</p><h3>The why underneath all of this</h3><p>There are very few people using AI to access their inner voice, calm their nervous system, and create from that place.</p><p>Most of the AI conversation right now is about speed. Efficiency. Scale. And I get it. Those things matter. But they’re the “what.” And the “what” is easy to talk about.</p><p>The “why” is harder. The “why” is that I’m a human being whose life got very hard, very fast, and I needed a way to keep creating without abandoning the part of me that knows what’s true. I needed to remove the friction of survival so my inner voice could actually be heard. Not so I could scale a content operation. So I could hear myself think.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1Siddun7dBip8EicXjUb9g.png" /></figure><p>Prompt engineering, for me, is not about getting AI to produce output faster. It’s about learning to describe the feeling of a regulated nervous system precisely enough that a machine can help you stay there while you do the hard work of building a life.</p><p>That’s what I’m building toward. Not “use AI to post more.” Use AI to access the part of yourself that already knows what to do. And then create from there.</p><h3>I’m still in the middle of this</h3><p>If you’re in a season where the friction feels like it’s winning, where you have so many ideas and so little capacity to actually do them, I want you to know something. The way through isn’t more discipline. It isn’t a better to-do list. It isn’t grinding harder.</p><p>The way through, for me, was starting with the body. Asking what regulation feels like. Building a system that reflects that feeling. And then letting the technology meet me where I actually am instead of where I think I should be.</p><p>I’m not writing this from the other side. I’m writing it from the middle. But I’m writing it. And that’s something.</p><p>If you’re in the middle too, I’d love to know what’s working for you.</p><p><em>This is the second in a series about building with AI as a non-technical founder. The first, </em><a href="https://medium.com/@eleahminer/what-my-ai-team-actually-does-ecdad70a2375"><em>What My AI Team Actually Does</em></a><em>, covers the architecture. This one covers the why.</em></p><p><em>I’m building </em><a href="https://leala.co/"><em>LEALA</em></a><em>, a somatic healing and inner voice community. Everything I’m learning about AI, nervous system regulation, and creating from alignment gets shared in real time.</em></p><p><a href="https://www.linkedin.com/in/eleah-miner-844969b8"><em>LinkedIn</em></a><em> · </em><a href="https://www.instagram.com/leala.io"><em>Instagram</em></a><em> · </em><a href="https://www.youtube.com/@leala-expansion"><em>YouTube</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=09bd865fee6e" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>