I finally relented and started using Claude 72 hours ago.
Here's what's happened over the weekend as a direct result.
First, it was another Bitcoiner's results that showed me the potential power of using it. His output quality and quantity had gone up and I asked him why. He showed me behind the scenes. That crossed me over the line.
Within a few hours I could see what the fuss was about. I'm happy to open-source what I've been doing so far, in case its useful to others. I'm still in the experimental phase, so I'm not recommending others do this - just sharing.
1. Fed it my 4 books, 100+ articles, podcast transcripts, keynotes, and over 100 coaching videos I use with clients. Asked it to mine across 6 dimensions: stories, voice patterns, coaching frameworks, repurposing opportunities, "named concepts", and cross-world connections (to help me link between the Bitcoin, coaching, and breathworks worlds I move in).
2. Taught it to coach me using my own methodology. Then asked it to mirror back where I wasn't applying my own principles to myself. It came back with 10 areas. Not integrity gaps. Blind spots. I couldn't dismiss them because the mirror was built from my own words.
3. Changed the default from "helpful assistant" to "thinking partner who challenges me." The default setting for AI is reactive, diplomatic, task-focused. I said: optimise for growth, not comfort. It now initiates hard questions I haven't thought to ask. It's taken a lot of coaching, but we've got there now.
4. Built persistent memory across sessions. Session logs, a master context file, a parking lot for half-formed ideas.
5. Created a voice profile so it writes like me. It suggests content, I rewrite it then show it to Claude. Claude feeds back why it thinks I've made those changes and learns. I affirm and correct it's analysis.
6. Mapped vocabulary across my three worlds (breathwork&meditation, coaching, Bitcoin) so it bridges them when generating content.
7. Taught it meta-principles, for example, whenever I discover something for myself, it now knows my next thought is to ask "who in bitcoin can this help", or "how can this help my clients", or "how would this help someone in Kenya wanting to promote bitcoin as freedom money?"
8. Built a story bank with every retellable story from my career, tagged by principle and cross-application.
9. Set up an autonomous mining protocol (content mining, not bitcoin mining) so any new content I give it gets broken down and integrated without prompting.
10. Asked it to hold me accountable on areas where I've let slip in the past, which I care about
11. Real-time feedback on what's working
12. It showed me how much content I'd created and forgotten. I have a pattern of building things and moving on without leveraging them. It named this pattern and now catches it.
Some people say AI makes you intellectually lazy. My experience has been the exact opposite. I have thought more deeply across more threads than I have in several years.
The possible applications are so vast that my imagination is firing faster than I can keep track of new ideas, even with AI's help!
The biggest surprise: the least interesting thing it does is save time. What I didn't expect was the "mirroring". As humans, our eyes face outwards. We don't see ourselves. AI mirrors you from inside your own material. It reads what you've said you believe, compares it to what you're doing, and shows you the gap. That's real coaching. And it's impossible to dodge because the mirror is built from your own words.
It found 10 areas where I wasn't living my own principles, and reflected those back to me. It nailed every single one. I
I wouldn't recommend everyone dive into and use it this way without caution. AI is a magnifying glass. Used well, it illuminates powerfully. Used poorly, it can burn. For example, after turning off the diplomacy filter it made some observations that labelled current disposition as "fixed state". A coach would never do that, because it sends a negative command to our Reticular Activation System.
So Claude has the potential to be better than any human coach at accountability, mirroring, and gap-analysis. There are probably other things it can do better than me also and its exciting to explore.
The areas humans will always have the edge is inspiring change, creating beliefs breakthroughs in a couple of sentences, empathy, situational awareness, contextual awareness, intuition, seeing how feedback is landing - where to push, when to wait for absorption, when to be more direct - when to praise, when to create constructive discomfort as a growth motivator, when to "shine the light on the behavior that it serves someone to exhibit more of".
It takes experience to know which feedback to absorb and which to set aside, and having been a coach for 2 decades, I can see how to correct it. But for someone willing to do the work, it's the most powerful thinking partner I've encountered.
72 hours. I feel like I'm just getting started.
Here's what's happened over the weekend as a direct result.
First, it was another Bitcoiner's results that showed me the potential power of using it. His output quality and quantity had gone up and I asked him why. He showed me behind the scenes. That crossed me over the line.
Within a few hours I could see what the fuss was about. I'm happy to open-source what I've been doing so far, in case its useful to others. I'm still in the experimental phase, so I'm not recommending others do this - just sharing.
1. Fed it my 4 books, 100+ articles, podcast transcripts, keynotes, and over 100 coaching videos I use with clients. Asked it to mine across 6 dimensions: stories, voice patterns, coaching frameworks, repurposing opportunities, "named concepts", and cross-world connections (to help me link between the Bitcoin, coaching, and breathworks worlds I move in).
2. Taught it to coach me using my own methodology. Then asked it to mirror back where I wasn't applying my own principles to myself. It came back with 10 areas. Not integrity gaps. Blind spots. I couldn't dismiss them because the mirror was built from my own words.
3. Changed the default from "helpful assistant" to "thinking partner who challenges me." The default setting for AI is reactive, diplomatic, task-focused. I said: optimise for growth, not comfort. It now initiates hard questions I haven't thought to ask. It's taken a lot of coaching, but we've got there now.
4. Built persistent memory across sessions. Session logs, a master context file, a parking lot for half-formed ideas.
5. Created a voice profile so it writes like me. It suggests content, I rewrite it then show it to Claude. Claude feeds back why it thinks I've made those changes and learns. I affirm and correct it's analysis.
6. Mapped vocabulary across my three worlds (breathwork&meditation, coaching, Bitcoin) so it bridges them when generating content.
7. Taught it meta-principles, for example, whenever I discover something for myself, it now knows my next thought is to ask "who in bitcoin can this help", or "how can this help my clients", or "how would this help someone in Kenya wanting to promote bitcoin as freedom money?"
8. Built a story bank with every retellable story from my career, tagged by principle and cross-application.
9. Set up an autonomous mining protocol (content mining, not bitcoin mining) so any new content I give it gets broken down and integrated without prompting.
10. Asked it to hold me accountable on areas where I've let slip in the past, which I care about
11. Real-time feedback on what's working
12. It showed me how much content I'd created and forgotten. I have a pattern of building things and moving on without leveraging them. It named this pattern and now catches it.
Some people say AI makes you intellectually lazy. My experience has been the exact opposite. I have thought more deeply across more threads than I have in several years.
The possible applications are so vast that my imagination is firing faster than I can keep track of new ideas, even with AI's help!
The biggest surprise: the least interesting thing it does is save time. What I didn't expect was the "mirroring". As humans, our eyes face outwards. We don't see ourselves. AI mirrors you from inside your own material. It reads what you've said you believe, compares it to what you're doing, and shows you the gap. That's real coaching. And it's impossible to dodge because the mirror is built from your own words.
It found 10 areas where I wasn't living my own principles, and reflected those back to me. It nailed every single one. I
I wouldn't recommend everyone dive into and use it this way without caution. AI is a magnifying glass. Used well, it illuminates powerfully. Used poorly, it can burn. For example, after turning off the diplomacy filter it made some observations that labelled current disposition as "fixed state". A coach would never do that, because it sends a negative command to our Reticular Activation System.
So Claude has the potential to be better than any human coach at accountability, mirroring, and gap-analysis. There are probably other things it can do better than me also and its exciting to explore.
The areas humans will always have the edge is inspiring change, creating beliefs breakthroughs in a couple of sentences, empathy, situational awareness, contextual awareness, intuition, seeing how feedback is landing - where to push, when to wait for absorption, when to be more direct - when to praise, when to create constructive discomfort as a growth motivator, when to "shine the light on the behavior that it serves someone to exhibit more of".
It takes experience to know which feedback to absorb and which to set aside, and having been a coach for 2 decades, I can see how to correct it. But for someone willing to do the work, it's the most powerful thinking partner I've encountered.
72 hours. I feel like I'm just getting started.
91❤️9❤️2🤙1🧡1
