That small non-decision turned into an hour of thinking about what I've been doing to myself.
The logical brain is useful. Mine has been running at full capacity for months, and I can point to what it's built: a leveraged portfolio system that manages itself while I sleep, an AI assistant I've written about here, a morning briefing that synthesizes my reading automatically, a blog with actual posts in it. I'm doing a book club with friends, reading the picks, showing up to the conversations. I'm doing full-time parenting while Inwu works, which is its own entire job with its own planning and logistics and emotional bandwidth. I'm learning technical analysis seriously, building out the trading system, keeping up with the reading that feeds it.
None of that is bad. All of it is meaningful to me. But when I listed it out in my head at the gym this morning, I noticed something: every single item on that list is for something. The portfolio system is for financial freedom. The book club is for my friendships, and also, honestly, for keeping my brain sharp. Being a good father is for Flynn. The trading is for the family's future. The blog is for building something that lasts. Every hour I spend is accounted for, pointed at a goal, optimized toward some version of a better life.
There is nothing on that list that is just for me, in the way that music is just for me. Not useful, not growth, not investment in anything. Just a feeling.
And I had tricked myself into believing that was the right way to live.
The trick I'd been running on myself goes like this: music is fine, but if you're going to put something in your ears anyway, why not make it count? A podcast about finance. An audiobook on stoicism. Something that feeds back into the machine. The logic is airtight. The music takes about the same time and leaves you with less to show for it.
The problem is that "less to show for it" is exactly the point. The part of you that needs music isn't looking for a return on investment. It doesn't care about compounding. It just wants to feel a chord change land, or get lost in a lyric for a second, or have some piece of sound connect to something in your chest that you'd forgotten was there. That part of you doesn't speak the language of optimization. And if you run your inner life entirely in that language, it goes quiet. It doesn't fight back, it just... recedes. Until one day at the gym it surfaces for a moment and the feeling is almost grief.
Since finishing Spark there's been a small void, which is probably why I've been filling evenings with Slay the Spire 2. It just came out, it's good, and it scratches the itch of having something to work on. But it's still a thinking activity: you're reading cards, considering synergies, making decisions every few seconds. My brain never actually stopped. Music was the first thing I'd done in a while that asked nothing of me at all, and I think that's what made the feeling so surprising.
I wrote a few weeks ago about trying to be like a tree: rooted, unswayed by things that don't matter, strong enough to shelter the people around you. The idea was about building a centre that doesn't move when everything else does. What I hadn't thought about was what that centre needs to be rooted in. If it's only rooted in productivity and goals and optimized output, it's still a kind of reactivity. You're just reacting to an ideal of yourself instead of to the chaos outside you.
A tree doesn't grow in order to grow. It grows because it's alive.
The ache I felt wasn't nostalgia, exactly. It wasn't longing for a specific time in my life when I listened to more music. It felt more like a signal from a part of myself I'd stopped listening to, the part that doesn't have a five-year plan, that doesn't fit cleanly into the project of becoming who I want to be. That part isn't irrational. It's just not amenable to being scheduled and justified.
And I think I've been afraid of that part, a little. Because if I let it have time, I can't point to what it produced. If I sit with music for an hour, I don't have a new mental model or a finished task or a stronger relationship to show for it. My logical brain doesn't know what to do with that. So it quietly reroutes me toward something measurable, and I follow along because the logic is always sound.
The logic is always sound. That's the problem.
I came across a Substack post recently that named something I'd been circling without landing on. The argument was simple: there's a difference between an optimized life and an interesting one, and the two are in tension in a way most productivity culture doesn't want you to notice. Systems are useful when you already know what you're building. When you don't, they become a substitute for figuring that out. A man who doesn't know what he's building uses optimization as a way to feel like he's moving without having to decide where.
Interesting people are interesting because they've accumulated texture that can't be scheduled. They pursued things because those things pulled them, not because they compounded. They made choices that were genuinely theirs, not because the logic was airtight but because they wanted to. The writer's line near the end stuck with me: the goal was never a better routine. The goal was a life that made the routine irrelevant.
That's the version I'd been losing track of. I had the routines. I had the systems. And the life inside them was starting to feel a little absent.
I don't have a system for fixing this. My first instinct, genuinely, was to figure out how to build more joy into my life: optimize my happiness, track my free time, add "listen to music" to a habit app, check it off once a day. I caught myself mid-thought and realized that's the entire problem in miniature. The logical brain, handed a feelings problem, immediately tries to schedule it into submission.
What I actually need is simpler than that. I need to want something and do it, without it being for anything else. Put on music and let it be music. Go for a walk without turning it into a thinking walk. Read something that has nothing to do with anything I'm building. Not as a technique. Just because I want to.
I don't know how consistent I'll be. The pull toward useful things is strong, and my life is genuinely full. But I felt something real this morning and I'm not ready to let the logical brain talk me out of it again.
The music is still going and I'm going to let it. Because I want to.
]]>The problem was that it never really delivered on the promise. Searching the web sounds impressive until you realize the results are inconsistent and the reasoning on top of them is unreliable. It made more mistakes than useful responses, the API costs were getting expensive even for short one-off requests, and the biggest issue was that it had no memory at all. Every conversation started from zero. So I ended up not using it much and went back to my regular stack: subscriptions to ChatGPT and Claude, cycling between them depending on the task.
A few weeks to a month ago, I came across OpenClaw, a bot that was taking the AI community by storm. The core idea was clever: it runs locally on your computer, uses a coding agent to manage a bunch of files as persistent memory, and you access it through a gateway on your phone. So it was basically Claude Code running continuously, but with a phone interface and a single giant repository it could manage and control over time. You could even hand it access to your entire computer. It felt like the first time persistent memory actually worked the way it should.
I gave it a try and genuinely loved it. The conversations felt coherent in a way nothing else had. But I could see the problem almost immediately: it was consuming my Claude subscription credits at an alarming rate. OpenClaw defaulted to Opus, Anthropic's most capable and most expensive model, for everything. Every request, no matter how simple, would trigger a reasoning loop that burned through tokens like it had something to prove. There's a well-known example in the community of someone saying "hi" to an OpenClaw instance and watching it spin up a context-heavy multi-step process that cost nearly $20. For a single greeting.
The math was ugly. Running it the way I wanted would have cost over a thousand dollars a month, which is five times the cost of my most expensive Claude subscription. Anthropic eventually banned OpenClaw from their platform entirely. I believe it was the last straw that led them to restrict subscription usage to Claude Code projects only, since OpenClaw was essentially jailbreaking the subscription model. Smart move from them, and honestly an overdue one.
I'd already named my OpenClaw instance Spark, and when OpenClaw got shut down, I decided to keep the name and build the real thing myself.
The two things I wanted to get right were cost efficiency and memory. OpenClaw's single-model-for-everything approach was the root of both problems: wasteful because it used Opus for tasks that didn't need it, and brittle because the memory was just a pile of files without any real structure. I'd also spent enough time building with MCPs (model context protocol tools) to know they're unreliable. They use a lot of credits, frequently return bad results, and break in hard-to-debug ways. I didn't want to build on top of that.
So I built my own action system for Spark instead, and a memory system designed around how conversations actually work. Information within a single conversation stays cohesive and contextually linked. Information from past conversations gets pulled in automatically through semantic search, so if you mention something that's relevant to what Spark already knows about you, it surfaces that context without you having to ask for it. It just shows up in the conversation, the way it would with a person who actually remembers your history.
The design goal for Spark's memory was that it should work the way human memory works. You don't consciously retrieve memories by querying a database. You're in a conversation, something triggers a connection, and the relevant context floats to the surface. That's what I wanted Spark to do.
When you're talking to Spark and you mention something it's encountered before, it brings that context forward automatically. You don't have to say "remember when we talked about X." It just knows, and it acts like it knows. The first time this happened naturally in a real conversation it felt like talking to an old friend rather than a tool.
The system is also built to run for years without falling apart. The memory doesn't degrade. It doesn't run into context limits. It's not tied to a single conversation thread that gets dropped when you close the app. It's persistent in the way that actually matters.
The other core design principle was to never use a single model for everything. This was OpenClaw's biggest architectural mistake. Opus is brilliant for hard reasoning problems, but it's grotesquely expensive for simple tasks, and most tasks are simple. Routing everything through your most powerful model because you haven't thought about the problem is a sign that you're building carelessly.
Spark routes by task type. For general conversation, I've landed on Grok as the best fit. Its style is concise and direct in a way I find genuinely useful, probably a product of being trained heavily on Twitter where every word has to earn its place. For coding, Spark hands off to a completely separate environment running ChatGPT Codex. That environment shares context with the main conversation so it knows what we've been discussing, but the code work happens in its own space and doesn't bleed back into the main chat. Keeping them separate means the main conversation stays clean and readable regardless of how deep a coding session gets.
Context trimming was the other piece I had to get right. Long conversations accumulate a lot of noise: half-formed ideas, resolved questions, tangents that went nowhere. If you don't actively prune that, the model eventually loses the thread on what actually matters. Spark continuously trims its active context down to the most important information, so the conversation stays fluid even over long sessions without losing the substance of what we've covered.
Once I had the conversation side working the way I wanted, I needed Spark to actually run as infrastructure, not just a script I fire up manually. I built it on a Uvicorn server so it runs persistently on my computer. This matters more than it sounds. Cron jobs are brittle for anything complex: they run at fixed intervals, they don't communicate with each other, and when something goes wrong you usually find out hours later. Spark's scheduler can handle real complexity, running other projects on consistent schedules, managing dependencies between jobs, and sending me outputs from any task that finishes. It's closer to having a process manager than a timer.
Spark is the assistant I actually wanted when I first built that N8N bot over a year ago. It knows me, it remembers what we've talked about, it doesn't cost a fortune to run, and it handles real work across conversation, code, and background scheduling.
It's not ready for anyone else to use yet. It's deeply customized to how I work, and there's a gap between "works for me" and "works for someone else" that I haven't bridged. But if this reaches anyone who's interested in what I've built here, either to learn more about the architecture or to help figure out how to make something like this available more broadly, I'd genuinely like to hear from you. The goal was always to build an AI partner that could actually help manage my life for years. I think Spark is becoming that. Here's to building AI tools that actually work.
]]>I'd describe the idea to Claude, get something back that was structured and coherent and said all the right things, and then I'd sit there trying to figure out why it didn't feel like mine. The arguments were sound. The flow was fine. But it read like a blog post about taste written by a system that doesn't have any. I kept editing, pushing it away from the centre, trying to find the version that actually sounded like me thinking out loud rather than an AI summarizing a concept.
That editing process is the whole point of this post.
Every model you interact with is, at its core, a compression of human knowledge. All of it, every book, every conversation, every forum post, every research paper, averaged and weighted into something coherent. That's remarkable, and it's also the limitation hiding in plain sight.
Inwu and I have been watching Pluribus, the Vince Gilligan show where an alien virus unifies almost all of humanity into a single collective consciousness. The "Others," as they're called, are peaceful, content, and have access to the combined knowledge of every person on Earth. They're also incredibly boring. Inwu pointed it out first: they talk like ChatGPT responses. Coherent, reasonable, and completely devoid of anything interesting. That's what happens when you merge all of humanity into one mind, you get the average of everyone, which is the voice of no one in particular.
AI works the same way. Ask it to write and you get competent, median-quality prose. Ask it to strategize and you get the advice that would survive a committee review. All knowledge at your fingertips, and none of the sharp edges that make ideas interesting. I use AI constantly, and the flatten-everything capability is exactly what makes it powerful for execution, but when everyone has access to the same flattened intelligence, the differentiator isn't access anymore, it's direction.
I wrote recently about how essentialism changes when creation becomes cheap. The old constraint was effort: you couldn't build everything, so you had to choose carefully. Now AI removes that constraint, and the scarcity shifts from "can I make this?" to "should I?"
That post was about saying no. This one is about what's left when you do: taste. The idiosyncratic, personal, hard-to-articulate sense of what matters. The "I don't know why but this feels important to me" instinct that doesn't come from data or consensus but from the accumulation of everything you've experienced, filtered through whatever it is that makes you specifically you.
AI can't replicate that, not because it isn't sophisticated enough, but because taste is definitionally non-consensus. It lives at the edges of the distribution, exactly where the averaging process loses signal.
In business, a moat is what keeps competitors out. Traditionally that meant patents, network effects, economies of scale. But AI is dissolving execution-based moats fast. If a model can write your code, design your interface, draft your copy, and analyze your market, then the things that used to take teams of specialists are table stakes.
What remains is the weird stuff, the specific combination of interests and obsessions that led you to see something nobody else saw. The willingness to pursue an idea that looks wrong on paper because something in your gut says it matters. The accumulated context of your particular life, your failures, your experiments, your late-night rabbit holes that didn't obviously lead anywhere until suddenly they did.
My portfolio system is a good example. It didn't come from asking an AI for the optimal strategy. It came from years of reading, backtesting, and filtering everything through my own sense of what I could actually stomach in a bad quarter. The optimal strategy doesn't have edges, because it's what everyone would build, which means it's what no one has an advantage building. The interesting stuff lives in the specific choices that only make sense given your particular context.
I keep coming back to essentialism because the concept keeps evolving as AI gets more capable. When I wrote about it before, the insight was that the skill shifts from doing to deciding. But there's another layer underneath that.
Knowing what not to do becomes the entire game when AI can do anything you ask, and "knowing what not to do" isn't just discipline or prioritization in the traditional sense. It's having a strong enough internal compass that you can look at a field of infinite possibilities and feel which direction is yours.
This is why I think capital allocation is the right mental model for the AI era. Not capital in the narrow financial sense (though that applies too), but capital as your finite resources: time, attention, energy, money. You become the allocator, the AI executes, and your job is to point it somewhere worth going. "Somewhere worth going" can't be optimized because there's no objective function, there's only what matters to you.
My daily workflow with Claude is a live example. Claude has access to vast knowledge and can execute at a level I couldn't match alone, but the system only works because I bring the context and direction. I know what I'm trying to build, why it matters, what trade-offs I'm willing to accept, which of the ten possible approaches fits my specific situation. Without that direction, AI produces excellent average work. With it, you build things that are yours in a way that matters.
This connects back to distilling to the bones and rebuilding with intention. You strip a problem to its core, apply your particular lens, and let the AI handle the reconstruction. The taste is in the stripping and the lens, and the execution is the part that's rapidly becoming free.
If taste is the moat, the practical question is how you develop it. I don't think there's a formula (if there were, it could be automated), but I've noticed patterns in how it works for me: reading widely but following my curiosity rather than a curriculum, paying attention to what I find myself drawn to when nobody's watching, building things for myself before building for an audience, having strong opinions and holding them long enough to actually test them against reality before updating honestly when reality pushes back.
The common thread is that taste develops through contact with the world, not through optimization. You can't A/B test your way to an interesting perspective, you have to live one.
Inwu has this instinct naturally. When I was deep in building Liquid Notes, she didn't tell me the app was bad. She said it was great, but maybe it wasn't my calling, that my unique skills pointed somewhere else. She was seeing something I couldn't see because I was too close to the work: not whether I could build it, but whether it was the right thing for me specifically to be building. That's taste applied to someone else's life, and she was right.
And the most important part is being willing to be wrong in interesting ways rather than right in boring ones. The edges are where the interesting mistakes live, and those mistakes are what eventually become your specific, non-replicable point of view.
The people who thrive in the next few years won't be the best executors, because execution is being commoditized rapidly. They'll be the ones with the clearest sense of direction, the ones who can look at everything AI makes possible and say "this, not that" with conviction.
The moat isn't in knowing how, it's in knowing why. And "why" is personal, messy, unoptimizable, and entirely yours. Build it by living a life worth having taste about, and let the AI handle the rest.
]]>When musicians record a demo, they often fall in love with it. The specific sounds, the particular arrangement, all the little details that made it feel alive in that moment. But those details can become a cage. You can't expand a completed version when there's no room to grow.
So what do some of the best do? They write the demo out as sheet music. Strip it to the skeleton: just the melody, the chords, the structure. Destroy everything that isn't essential. Then hand it to the band and rebuild from the bones.
The result isn't a polished version of the original. It's something new, something with room to breathe.
This connects to something Charlie Munger talks about: inversion. Instead of asking "how do I make this better?" you ask "what would make this worse?" and then avoid those things. Don't try to be brilliant. Just don't be stupid.
The instinct for most builders is to add. When something isn't working, the question is usually "what can I add to fix it?" But sometimes the better question is "what can I remove?"
Not every problem needs a new feature. Sometimes the answer is fewer moving parts, not more.
There's a short document called the Cult of Done Manifesto, written by Bre Pettis and Kio Stark in 2009. It has thirteen principles about finishing things, and principle eleven says simply: "Destruction is a variant of done."
That reframes everything. Throwing something away isn't failure. It's completion. Tearing something down to its core and starting fresh isn't going backward, it's a form of finishing. You're done with the version that no longer serves you.
The manifesto also says "done is the engine of more." Completion isn't the end, it's what creates space for the next thing. And sometimes the fastest way to complete something is to destroy the parts that are weighing it down.
The real skill isn't building. Anyone can add. The skill is knowing what's load-bearing and what's just there.
Systems want to expand. Features accumulate. Complexity creeps in because each addition made sense at the time. But eventually you're maintaining things that don't matter anymore, and the weight of all that maintenance slows everything down.
This is especially relevant now that AI can generate and expand almost anything on demand. The bottleneck isn't creation anymore. It's curation. Knowing what to keep and what to cut. Having an eye for the essential.
Expansion is easy. Distillation is hard.
The Rubin insight isn't about destruction for its own sake. It's about creating room. You strip something to its skeleton not to abandon it, but to rebuild it better. The melody is still there. The core is still there. But now there's space to grow in directions the original version couldn't accommodate.
If you're a builder, someone who loves adding layers and refining systems, this might feel counterintuitive. The instinct to add is strong because it feels like progress. But sometimes the most productive thing you can do is subtract. Strip something back to its bones, see what's actually essential, and only then decide what deserves to be rebuilt.
What have you built that might need breaking down?
]]>I'd cobbled together something close using Obsidian and various plugins, but it never felt quite right. Then I discovered Claude Code, an AI that could actually write production software, and something clicked. I'd gotten pretty skilled at prompting. Maybe I could finally build the real thing, the app that matched my vision exactly.
So I started building.
It came together faster than I expected. Voice notes that transcribed and condensed themselves. A zoom slider that moved smoothly between single keywords and full paragraphs. Offline support so it would work at the gym where the WiFi blocks half the internet. A progressive web app that ran on my phone without going through any app store.
Every day I'd work on it while Flynn napped, adding features, fixing bugs, watching my vision materialize on screen. I learned how to host a website, how to manage domains, how to forward emails, how to optimize for mobile. I learned actual web development, not just prompting but understanding what the code was doing and why.
By the end of two weeks, I had something genuinely useful. Something that worked exactly the way I'd imagined. My note app, running on my domain, built to my specifications.
And then Inwu asked me a question.
"What's the goal here?"
She wasn't being dismissive. She's never dismissive about my projects. But she's precise, and she saw something I was too deep in the building to see.
If I wanted this to be more than a personal tool, it would become a business. Taxes, support, hiring, marketing. A note-taking app, one of hundreds already fighting for attention in a market where AI plus notes is literally everyone's first idea. There are so many of them now. They're all trying to do this.
Meanwhile, I'd spent the past year building something else: a trading system. An AI council that debates market positions. Algorithms that manage my capital while I sleep. That system has already doubled my money. It runs almost autonomously. There are no users, no support tickets, no company to build. Just me and my capital and an edge that keeps compounding.
Which one should I be spending my time on?
My financial skills are rare. Trading edges are scarce, and I actually have one. Note-taking apps are crowded, and even the best ones struggle to make money. The answer was obvious once I stopped to look at it.
And there's something else. Flynn is nine months old. I'm on parental leave until next fall. This window, where I can watch him discover how blocks stack and cheer when he pulls himself up on the furniture, doesn't reopen. He'll never be this age again, and I'll never get this time back.
Every hour I spend debugging a UI quirk is an hour I'm not present with him. Every feature I add is a feature I'm choosing over my son.
This forced me to rethink what essentialism actually means.
Essentialism used to be about effort scarcity. You only have so many hours, so much energy, so much willpower. The discipline was doing fewer things because you couldn't do everything. Pick the vital few. Say no to the rest.
But what happens when doing becomes cheap? I can spin up a new feature in an afternoon. Claude writes most of the code. I provide direction, test it, iterate. The effort barrier that used to protect me from my own ideas has collapsed.
The scarcity didn't disappear. It shifted. Now the scarce resource isn't effort. It's attention. It's meaning. It's the finite capacity to care about things deeply enough to maintain them, to use them, to let them serve your actual life instead of becoming another thing you built and forgot.
The old discipline was "can I make this?" The new discipline is "should I?"
That's harder. Saying no to impossible things is easy. Your limitations do the work for you. Saying no to possible things, things you could ship this afternoon, things that would be genuinely useful, that takes real clarity about what matters.
Two weeks of building wasn't wasted. I learned Claude Code and how to work with an AI on real software. I learned web hosting, domain management, mobile optimization, progressive web apps. I learned how to turn a vision into working software.
But the most valuable thing I learned was this: the skill isn't in the building anymore. AI has made creation trivially easy. You can imagine something and watch it materialize in days instead of months. The hard part now is knowing what not to build.
Liquid Notes works. It captures my voice notes, transcribes them, condenses them, syncs across devices. It does what I need it to do.
I'm not abandoning it. I'm declaring it finished. The difference matters to me. Abandoned means failure. Finished means the work is complete, not because there's nothing left to add, but because I've chosen to stop adding. The app works. It serves its purpose. And my time is better spent elsewhere.
My energy goes to the trading system now, the one that actually compounds. It goes to Flynn, who won't be nine months old again. It goes to being present with Inwu instead of debugging edge cases at midnight.
When you can build anything, the hardest question isn't "how?" It's "should I?"
And sometimes the answer is: I already have enough. Time to stop.
]]>Then she'll say it: "You should be more like a tree."
In Korean culture, there's an adage about the ideal man. A man like a tree is someone who doesn't sway with every wind. He grows slowly, steadily, rooted in place, strong enough to shelter the people around him. He's not cold or rigid, just unmoved by the things that don't matter so he can show up fully for the things that do.
I used to brush this off. That's not me, I'm the fun one. Trees are boring. But then I became a father, and I started to understand what she meant.
There's a moment that I think every new parent has, where you realize the chaos isn't going to stop. The sleep deprivation, the crying, the constant low-grade anxiety of keeping a tiny human alive. You can't outrun it and you can't out-energy it. You can only learn to be still inside of it.
I've meditated on and off for years, but I never really took it seriously. It was something I did when I remembered, more of a box to check than a real practice. Becoming a father changed that. Not overnight, but slowly, in the hard moments. Flynn is crying, I'm exhausted, I can feel the frustration rising up in my chest and I want to react to it, I want to snap or shut down or reach for my phone. But I'm learning to do something different. I'm learning to feel that negative emotion come up and not associate with it. To step back from it, watch it, and let it pass without it becoming who I am in that moment.
I'm still not good at it. I'm taking meditation seriously for maybe the first time in my life, and I can feel the difference it's making, but it is still very difficult. Some days I sit there and my mind won't stop. Some days Flynn cries and I lose my patience anyway. But there's something growing in those moments of practice, even the failed ones. A centre, a quiet place that, when I can find it, doesn't move when everything else does.
That centre is your consciousness and your will. It's the part of you that knows what's right before you react, the decision-maker that, when it acts, you feel it in your chest. Not anxiety or excitement, something quieter. Alignment.
You have to grow it deliberately, the way you'd water a real tree. Daily, without drama, trusting that the roots are going deeper even when you can't see the progress.
Meditation grows it, but so does any practice where you choose stillness over reaction. Pausing before you respond when someone pushes your buttons, sitting with discomfort instead of rushing to fix it, holding your position when the market or life is screaming at you to move. Every time you choose the harder silence over the easier reaction, the trunk gets a little thicker.
I've been building this tree-like nature into my life as a core principle. It's become one of the things I care most about developing, not just as a practice but as a way of being.
This is where I think people get it wrong. Being like a tree doesn't mean you stop laughing or being excitable or being yourself. Trees bend in storms. They lose their leaves in winter. They're alive, constantly growing, constantly adapting.
The difference is where the movement comes from. A reactive person moves because the world moves them. Something happens and they get pulled along like a leaf in the current. A centred person moves because they choose to. The storm hits, the branches sway, but the trunk holds and the roots go deeper.
I still laugh too loud. I still get excited about things. Inwu still teases me for it. But there's something underneath now that wasn't there before, a weight, a centre of gravity that lets me be playful on the surface because I know I'm solid underneath.
My son Flynn is almost ten months old. He doesn't know what a tree is yet, and he doesn't know about Korean adages or meditation or personal philosophy. But he watches me, and he watches everything.
What I want him to see, not when he's old enough to understand words but right now when he's old enough to feel energy, is a father who is present. Not reactive or scattered or reaching for distractions when things get hard, but rooted. Steady. Growing.
Inwu was right all along. I'm just finally learning how to listen.
]]>One of YNAB's core principles is giving every dollar a job. This means allocating your money into different categories each time you get paid, similar to the envelope method. This discipline of allocating money at the beginning of the month and then spending it according to plan actually creates freedom. Immediately, I noticed that I no longer worried about my money or how to spend it. I could just check my app and see exactly how much I could spend in each category.
I created categories for everything: fun, groceries, car expenses, transportation costs, and more. Over the years, this process has become easier, and my worries about money have diminished.
I liken it to learning a martial art: the more you know, the more freedom you have in expressing that form. You don't have to think about throwing a punch; you just do it correctly. Similarly, understanding your money allows you to manage it intuitively.
I don't check my budgets every day anymore. I still review them periodically but can sense when I'm overspending or underspending. This has created freedom in my life. Even my wife, who was initially against budgeting, now sees it as liberating. Budgeting doesn't mean you can't have fun; it means planning how much fun you can afford each month.
Everyone should have a budget. I highly recommend checking out YNAB or exploring other budgeting apps that use similar methods like zero-based budgeting or the envelope method. (Full disclosure: the YNAB link is a referral link - if you sign up, we both get a free month.) Budgeting gives you control and freedom over your finances. It might be daunting at first, but once you get into the habit, it will undoubtedly improve your life.
]]>Hemingway believed a writer should only explain small parts of what's happening and let the reader's imagination fill in the rest. He's not telling you a story so much as drawing a map and trusting you to walk it yourself. The characters are described simply, directly (an old fisherman, a boy who cares about him), but that simplicity invites you to infuse your own meaning.
I've been thinking about creative destruction lately, about how you create a framework and then let someone else bring it to life. Hemingway does this with readers. He provides the bones of the story and trusts you to add the flesh from your own experience.
The old man's reflections on luck, on his past successes and long drought of catching nothing, kept connecting to trading for me. You're always searching for the big catch, navigating murky water, facing setback after setback while holding onto the belief that your skill matters even when luck doesn't cooperate. Hemingway never mentions trading. These connections were mine, pulled from my own life and laid onto his map.
At the end, when the boy sees the old man sleeping and starts to cry, Hemingway doesn't explain why. You just know. The old man dreams of lions on African beaches, a detail that recurs throughout the book, and in that final image you understand something about rest, about youth, about what a person carries with them even as everything else gets stripped away. No explanation needed. The meaning lands because you built it yourself.
This has applications beyond literature. In leading a team, you don't want to give every direction. You provide a blueprint, a small map that nudges people in the right direction, and let them fill in the details with their own skills and judgment. Over-specification kills creativity. People thrive when they have room to be themselves within a structure.
The Old Man and the Sea is one of those books that earns its reputation. Brief, spare, and somehow enormous.
]]>The core idea is clear: create a comprehensive, structured course that could deliver theory, test understanding, and adapt based on performance. I used Claude Code to set up the infrastructure: a folder structure, course theory document, and initial specifications.
The syllabus generation came first. I wrote custom prompts to produce a 180-day curriculum of 15 minute lectures organized in phases: Mechanics, Psychology, Strategy, and Macro knowledge. Each lesson got a title and short description in the syllabus. The length of 15 minutes as well as 180 days was deliberate. Trading has so many layers that you must work from the ground up and build upon each step slowly but surely, and I needed to be able to fit it into my busy fatherhood schedule so I split it up into daily "micro" lessons.
Then I built the specifications for the daily lesson generation itself. Each lesson needed to be TTS-ready. I wanted audio lectures I could consume while multi tasking, or walking with Flynn. After the lesson there would be five questions testing comprehension of the material. A rubric was also generated with specific grading criteria for those specific 5 questions. And finally a teacher prompt that would evaluate answers using that rubric and provide feedback to the user.
Claude Code generated the lessons one by one. The quality held up and as I tested I was surprised by how much I was learning, but I hit token limits fast. After two or three lessons, the system would stop working. Context windows filled up, and token consumption burned through my Claude Pro allocation quickly. I'm now generating lessons as I go rather than building the full 180 days upfront. If I develop this further, I am thinking that just-in-time generation every night for the next day might be more manageable.
The delivery system is simple, each morning I get a TTS lecture listen to it carefully and then submit answers through a voice recording that I send to a webhook using a custom tool. The system grades them, provides feedback, and if I've passed, queues up the next lesson for tomorrow.
The grading is the weakest part of the system. The feedback itself is detailed and useful, often pointing to gaps in my reasoning or places where I missed nuance. But the numerical marks cluster around 88/100 regardless of performance. Even when the written feedback is harsh, the score stays in that narrow band. I don't know why, but I have tried a few different things but I think there might be an inherent issue in LLMs where they have a hard time judging how evaluations fit int a scoring system. The rubric is specific, the criteria are clear, but something in the evaluation loop defaults to a good but not perfect score. I have noticed this in other projects too so this is not isolated.
Still, the system works at making me smarter. The lessons have already changed how I think about trading: how I read price action, how I understand risk, what I pay attention to in market structure and how I manage my own psychology.
Trading sits in a strange category. It's not exactly secret knowledge, but it's gated. No formal education path exists for retail traders. YouTube is full of people selling courses based on their personal approach, which might work for them but often doesn't teach you the underlying mechanics. The incentive structure doesn't favor genuine education.
AI learning systems like this are especially useful when traditional education doesn't serve you: where the knowledge exists but isn't organized into a curriculum, or when you need depth but don't have access to expert instruction.
Meanwhile I'll keep learning, and if this goes somewhere maybe one day I'll make it public.
]]>Traditionally, money was seen merely as a medium for exchange, but in today's world, it's much more than that. You don't just let it sit there. You can invest it, save it, change it, and make it work for you. When it's working for you, in essence, it's being burned to earn more. The idea isn't to hoard money but to create a feedback loop that gains you more.
Consider using money as a recursive resource; it can generate more of itself through compounding. This is one of the few human inventions that can create more of itself. While nature has its own forms of growth and reproduction, with money, you can gain a fixed amount of interest. It's a fascinating human creation.
Another way to think about it is as storing your time. You've externalized your fuel (your time and energy) into money. This externalized fuel can then be used productively. Saving money in an account where it's not put to work doesn't generate returns because banks simply hold it as is. However, placing it in a savings account starts to burn a little bit of it, giving you interest in return.
When you invest in mutual funds or ETFs, you're essentially burning the entire amount by giving it to someone else to use. This concept is what I call burning to return. You burn money to get more money back from that equation.
On the flip side is burning to use, which involves spending money on products or services. You get a physical object in return. People often focus on burning to use because they want immediate benefits. However, the whole point of burning to return is to get more resources that you can eventually use.
Hoarding money for its own sake is actually a trap. If you have all the money in the world but don't know how to use it or what you want from life, you'll feel empty. Money needs to be used effectively; otherwise, it's just sitting there without purpose.
The ultimate goal should be creating a burn-to-return loop where your investments yield more than your consumption requires. When this happens consistently, you'll reach a state where you no longer need to work, true retirement mode. Your principal remains intact while covering all your needs and wants through returns on investment.
In this scenario, you've saved enough fuel (money) to live out the rest of your days comfortably. You've contributed enough value to the world and have been rewarded with the gift of free time for the rest of your life.
Remember, money in itself does nothing; it's the burning and usage of money (your energy and time) that creates effectiveness. Always consider what you would use your money for and save for later if there's no immediate need. Time is valuable, and keeping this in mind will guide your financial decisions effectively.
]]>I'd been playing around with a strategy called "hedgefundie"1 in my Alpha portfolio for a while. It's a 55% UPRO (3x leveraged S&P 500) and 45% TMF (3x leveraged long-term treasuries) portfolio. The idea is that stocks and bonds are negatively correlated, so when one goes down the other goes up, and by leveraging both you get high returns with some built-in hedging.
But when I really dug into the backtesting2, I realized it didn't make it through the interest rate lowering environment very well. When rates dropped, those leveraged treasuries got crushed. I needed a different approach.
As I was researching alternatives, I realized I wasn't just managing one portfolio. Inwu and I had money scattered across multiple accounts: my Alpha trading account, our Nicola portfolios through CI Direct, some international exposure, cash sitting around. I'd been thinking about each one separately, but that's not actually how risk works. If I wanted to build the right overall profile, I needed to think about all of them holistically.
The CI Direct Nicola account was our safety net. It's been producing safe, uncorrelated returns for a long time, and I didn't want to mess with that. But in the Alpha portfolio, I could take more concentrated risks if I knew the rest of our money was stable. If I could figure out the right allocation across all the portfolios, I could get the risk-adjusted returns I wanted without having to actively trade or manage things constantly.
We had scattered portfolios and things I'd worked on individually, but we never created a holistic system until now.
I spent months backtesting different approaches. Most of this happened while Flynn was sleeping or playing with Inwu. I'd be running simulations in Excel, testing different leverage ratios, looking at historical drawdowns.
One of the things I discovered was that Bitcoin and gold add a layer of interest rate insensitivity that treasuries don't. Gold especially seemed to be what we needed, with bonds taking a back seat on leverage and filling out the portfolio in other places. This was a breakthrough. Instead of relying on the traditional stock-bond correlation that hedgefundie uses, I could build something more robust to different interest rate environments.
One of my core ideas was that I wanted to make this passive. I wanted to spend more time with Flynn and Inwu, not glued to a screen managing trades. This ties into my whole automation philosophy: put in a lot of upfront effort, build the system right, and then let it run.
Obviously I'll come back to review it, but I want to let this thing compound for years without constant tinkering. The system needed to be rules-based, systematic, and automated enough that the math tells me what to do rather than relying on gut feelings or market timing.
After all that research, I landed on a four-portfolio system that works together holistically:
The idea is that I can take concentrated risks in the Alpha portfolio (30%) while keeping the majority of our wealth (70%) in more stable, diversified holdings. The whole portfolio uses about 1.2x leverage: aggressive enough to boost returns, conservative enough to sleep at night.
I built an Excel spreadsheet that handles monthly rebalancing automatically. It tells me where to contribute new money based on which portfolio is most underweight. The system buys low and sells high without me having to think about it.
For the detailed breakdown of allocations, exposures, and the rebalancing mechanics, I wrote a separate technical post: My Leveraged Multi-Portfolio Strategy Breakdown.
I should be honest about the risks. When I ran backtests on the Alpha portfolio with its leverage and concentrated alternative bets, I saw significant drawdowns in historical stress scenarios, with 40%+ declines in bad markets. If that were my entire portfolio, I'd never sleep. But it's only 30% of our total allocation, and when you look at the whole portfolio, those drawdowns get substantially muted.
I'm a bit nervous about the leverage, I'll admit. But I'm willing to take the risks given how it all balances out on a grander scale across the portfolios. We have decades for this to compound. We both have stable incomes. We can weather drawdowns without panicking. And honestly, I can tolerate volatility in the Alpha portfolio because I know the full portfolio is balanced. Inwu can feel secure knowing 70% of our wealth is in traditional, diversified holdings.
Obviously my risk profile is higher than most, so note that too. Using 3x leveraged ETFs means you can lose money fast in volatile markets. You need a strong stomach for drawdowns, a long time horizon, an emergency fund outside the portfolio, and the willingness to stick with the strategy through pain.
Becoming a father changed my relationship with money. I stopped thinking about investing as an intellectual exercise and started thinking about it as a responsibility, a long-term plan that needs to work through decades of market cycles, life changes, and unknowns.
I feel extremely proud of this work that I was able to do while learning and taking care of Flynn. I did most of this in the times when he was sleeping and when he was playing with Inwu. The urgency of new parenthood gave me the push to build this. The discipline of systematic rebalancing will let it compound over time. And the balance between aggressive and conservative allocations lets me sleep at night.
This multi-portfolio strategy isn't perfect. No strategy is. But it's thoughtful, backtested, balanced, and automated enough that I can focus on what actually matters: being present for Flynn while knowing our financial future is on solid ground.
This is not financial advice. This is my personal strategy based on my specific situation, risk tolerance, and time horizon. Your situation is different. Do your own research, understand the risks, and consider working with a financial advisor.
If you're thinking about portfolio construction across multiple accounts, I'm happy to discuss approaches. Just know that leverage magnifies both gains and losses, so proceed with caution.
Hedgefundie's Excellent Adventure - A detailed explanation of the original UPRO/TMF leveraged portfolio strategy ↩
I use Portfolio Visualizer for backtesting portfolio strategies and analyzing historical performance ↩
This post covers the specific allocations, exposures, rebalancing mechanics, and backtest results.
I manage our household wealth across four portfolios that work together as a single holistic system:
Quick note: I do also have a small trading account where I actively trade with money I'm okay losing. That's completely separate from this structure and I might write about that approach in a future post. This post is about our core passive portfolio strategy.
The Alpha portfolio (30% of total household wealth) is where I use leverage and alternative assets. Even though I call it "Alpha," this is a passive strategy. I'm not actively trading or timing the market. I developed a rules-based allocation that I hold and rebalance systematically.
Within the 30% Alpha allocation:
Yeah, those percentages add up to more than 100%. That's the leverage working.
In terms of total household portfolio exposure, this 30% slice gives us:
The Nicola Core is the stability engine at 40% of our household wealth. These are Nicola portfolios that we access through CI Direct's balanced private portfolio (split between Inwu and me). Normally, Nicola requires a $1 million buy-in, but CI Direct provides access through their balanced private portfolio, which I really appreciate. It's a great portfolio with a slightly higher fee than using Nicola directly, but still worth it.
The portfolios are conservatively allocated across:
At the 40% portfolio weight, this gives us:
The World Custom portfolio at 20% gives us exposure to markets and assets that neither of the other portfolios fully covers:
When you roll it all up across the four portfolios, our total household exposure looks like this:
You'll notice these add up to more than 100%. That's the leverage in the Alpha portfolio at work. We're using about 1.2x leverage at the total portfolio level: aggressive enough to potentially boost returns, conservative enough to sleep at night.
I built an Excel spreadsheet to track all of this and handle the rebalancing. Every month, the sheet calculates the current value of each portfolio, the actual allocation percentages versus the targets, and how much each portfolio has drifted.
I use a one-twelfth partial rebalancing approach rather than full rebalancing. This means I rebalance 1/12th of the drift each month rather than completely resetting to targets. It's a bit more effort than quarterly rebalancing but still passive and follows the system.
When we have money to invest, I don't just split it evenly. I contribute to whichever portfolio is most underweight relative to its target:
The system naturally buys low and sells high through contribution patterns.
Once a year, I review the portfolio-level allocations to see if we need to adjust the top-level targets (the 30/40/20/10 split) based on:
I backtested this strategy using Portfolio Visualizer and ran multiple historical scenarios.
When I ran simulations on the Alpha portfolio with its leverage and concentrated alternative bets, I saw significant drawdowns in historical stress scenarios, with 40%+ declines in bad markets. If that were my entire portfolio, I'd never sleep.
But it's only 30% of our total allocation. When you look at the whole portfolio, those drawdowns get substantially muted:
Over the long run (and we're talking 20+ year time horizon here), the backtests showed that the leveraged Alpha exposure added meaningful returns while the portfolio-level diversification kept total drawdowns within acceptable ranges.
The system works because we're not going all-in on leverage. We're using it strategically in one portfolio while maintaining balance across the whole.
The Excel system removes emotion from the process. The math tells us where to contribute. We don't have to guess or time the market. And if life changes (job loss, health issues, different risk tolerance), we can adjust the top-level allocations annually without changing the underlying strategy.
Using 3x leveraged ETFs means you can lose money fast in volatile markets. Using 2x gold means you're amplifying an already volatile asset. You need:
I'm using leverage in 30% of the portfolio because I believe the long-term expected returns justify the risk, and because the other 70% provides enough stability that we can ride out the storms. Obviously my risk profile is higher than most, so note that too.
If you're considering leverage, backtest it. Understand the drawdowns. Make sure you can actually stick with it when your account is down 35% and everyone is panicking.
This is not financial advice. This is my personal strategy based on my specific situation, risk tolerance, and time horizon. Your situation is different. Do your own research, understand the risks, and consider working with a financial advisor.
For questions about the philosophy and journey behind this strategy, see Building a Strategic Portfolio System as a New Parent.
]]>Over the year, I've accumulated hundreds of millions of tokens across different models as I upgraded, experimented, and built systems that work. That's roughly equivalent to writing 4,000 novels worth of text. All this testing and development has taught me deeply about how AI models work together, what they're each good at, and what it really takes to build reliable AI systems.
Using more credits than 98% of people isn't about running massive queries for no reason. It's about building systems that compound value every single day. Here's what's generating all those tokens:
A Daily Newsletter That Actually Gets Read: Every morning, I process multiple newsletters, my personal notes from the previous day, and market data into a single coherent briefing. It's personalized to my interests and includes perspectives from a "council" of AI advisors who challenge my thinking. Six months in, I've used it every single day, and it's become an essential daily augment to how I start my morning.
A Trading Recommendation Engine: This is by far my heaviest usage, with around 80% of my credits going here. The system has multiple layers that process trading data including price data, news feeds, and sentiment data to generate the trading recommendations I use every single trading day. Getting this to work well required extensive testing with real credits, which taught me an important lesson: AI development is not cheap. You have to use real credits to test things out and iterate until the system actually works reliably.
A Custom Intelligence Chatbot: I've built a chatbot that integrates my task lists, calendar, web search capabilities, and personal knowledge base. These are very helpful systems that know exactly what you want and act in the way you'd like them to. The key advantage is that you can test and tweak them way more than any other bots available, making them truly personalized to how you work.
Note Summarization and Automation: I've built a custom voice capture tool that I use separately from OpenRouter since they only handle text at the moment. This tool extracts and sends data to OpenRouter APIs for collection and structuring. Throughout the day, I capture thoughts, meeting notes, article highlights, and random ideas. As I enter these notes, an automated system summarizes them, categorizes them, identifies action items, extracts key insights, and creates connections between related ideas. What used to be a messy collection of fragmented thoughts becomes a structured knowledge base I can actually use. These summaries feed directly into my morning briefing, creating a continuous loop where yesterday's insights inform today's thinking.
Each of these systems runs daily. Each one processes significant amounts of data. And together, they explain how one person ends up using more credits than 98% of people on the platform.
Different AI models are good at different things, and I've learned to orchestrate them rather than rely on just one.
Gemini Flash is incredibly fast and efficient for high-volume tasks. I use it for the heavy lifting: processing multiple newsletters, summarizing large documents, and handling repetitive analysis. It's not always the most nuanced, but for quick throughput and cost-efficiency, it's unmatched. I've also used Gemini Pro for tasks requiring large context windows and advanced intelligence.
Claude Sonnet is my number one choice for extremely important tasks. When I need something written well, when I need careful analysis of complex situations, or when I need AI that truly understands context and subtext, this is where I go. Quality, reasoning, and nuance are what I need, so this is where I spend my time when the task matters most.
Grok is really good with a different tone than other models. Its exclusive access to X (Twitter) data gives it an edge in certain areas, especially financial markets where many participants tweet their ideas and claims. This makes it particularly valuable for sentiment analysis and tracking market narratives.
GPT is good and handles basic tasks well, with its own unique voice that sets the standard others follow. However, I still prefer Claude's responses. They're better in almost every respect, more readable and human-like in their output.
I don't use one model for everything. I orchestrate them: fast models for processing, smart models for reasoning, current models for real-time data. Each one playing to its strengths.
What separates casual AI usage from using more credits than 98% of people isn't running a few big queries. It's building systems that generate value every single day.
My morning briefing alone has probably saved me 150+ hours over six months while making me better informed. My trading engine processes information I couldn't possibly analyze manually. My chatbot eliminates friction from dozens of daily decisions.
This is compounding automation. Each day builds on the last. I update the systems over time to make them better. Each interaction is more valuable because of all the previous ones.
Seeing these stats made me realize something uncomfortable: AI usage has an environmental cost, and using this many credits means I have a responsibility to acknowledge it.
Running AI models requires massive computational power. Those token counts represent real electricity consumption, real carbon emissions, and real environmental impact. The more I use these tools, the more I benefit from them, the more I need to be honest about that cost.
I've started researching carbon offset options, specifically looking at Offset AI, a service designed to offset the environmental impact of AI usage. I'm not sure yet if carbon offsets are the complete solution, but I know ignoring the problem isn't an option.
If we're going to build lives augmented by AI, if we're going to use it this extensively, we need to think about sustainability. Not as an afterthought, but as a core part of how we use these tools.
I do feel like I'm at the edge of something, and that excites me. I'm just an individual working at figuring this all out. These tools are fundamentally changing how we work and live, and I want to be on the front edge of understanding and building with them.
I didn't set out to use this many credits. But as I kept going, the use cases came to me. This lets me create things I never had the brainpower to build on my own. I basically have a team of people working for me in my mind, with each AI having a unique voice and skill set. It's almost like managing a team where you use the right model for the right use case, building systems that can work for you while you sleep, much faster than any human could.
The biggest lessons are that using the right model for the right task matters (efficiency and quality both count), that you need to build systems that create real value rather than just burning credits, and that you have to be honest about the environmental cost. AI development isn't cheap, and you have to use real credits to test and iterate until things actually work.
The token usage is just a byproduct of building systems that actually work. If you're thinking about how to use AI more effectively, focus on building systems that compound value over time. Find the tasks you do every day that could be augmented. Pick the right models for the right jobs. And be thoughtful about the resources you're using.
What systems could you build that would generate value every single day? What would your life look like if you had AI tools perfectly tuned to how you think and work? And how can we do this responsibly?
Let me know what you think. I'm always interested in how others are using AI to augment their lives.
]]>Like many people trying to stay informed, I'd accumulated subscriptions to every newsletter that promised to make me smarter about markets, productivity, and life. The problem wasn't the quality. Most of them were excellent. The problem was redundancy. Five different writers would cover the same news story, and I'd read essentially the same take five times.
I needed the insights without the repetition. So I did what any data analyst would do: I automated my way out of it.
I started simple. I set up a Power Automate flow to save all my newsletter emails to files. Then I created an LLM prompt that could summarize each one while keeping the unique voice and perspective of each writer. No generic summaries. I wanted to preserve what made each newsletter worth reading in the first place.
Then I wrote another prompt to combine all these summaries into a single, cohesive news update. Since I care most about markets, I added a specific focus there. What used to take an hour now took five minutes to generate and ten minutes to read.
I used this system for a few months and loved it. But then I realized I could take it further.
Why stop at newsletters? Every day I take notes: ideas, observations, things I want to remember. What if I could review yesterday's notes every morning as part of my briefing? I added that feature, and then I created what I call my "council of perspectives," different AI voices that give me advice on my notes and share their thoughts. One might approach things from a stoic philosophy angle. Another might focus on practical business strategy. Another on long-term thinking.
This council doesn't just repeat my thoughts back to me. They challenge them. They ask questions I didn't think to ask. They connect dots I didn't see. It's like having a morning conversation with advisors who've read everything I'm thinking about and want to push me further. The council often stimulates new discussions and ideas that I carry with me throughout the day.
I wasn't done yet. I added a section for daily dad advice, short, practical wisdom about fatherhood. And some quotes to reflect on. I also included a brief Vancouver news section since local context matters to me.
The whole thing comes together as a personalized briefing that's exactly what I need to start my day informed and inspired. I set up a macro to automatically send this briefing to ElevenLabs Reader, which converts it to audio. Now I get my custom morning update as a podcast, delivered fresh every day.
I've listened to my AI-generated morning briefing every single day for the last six months.
It's improved my life in ways I didn't expect. I'm more informed with less effort. I'm thinking about my own notes more critically. I'm getting perspectives I wouldn't have considered on my own. And I'm doing it all while I make my morning coffee and feed my son Flynn.
The beauty of this system isn't just the efficiency, though that's nice. It's the intentionality. I'm not passively consuming whatever lands in my inbox. I'm actively curating the information and perspectives that matter most to me, then having them delivered in exactly the format I need.
Building this system taught me that we don't have to accept the default way information comes to us. We can shape it, filter it, and combine it with our own thoughts to get perspectives that challenge us to grow.
The tools are there, the AI models are powerful enough, and the automation platforms are accessible enough. What's missing is just the willingness to spend a weekend building something that will save you hours every week and make those hours more valuable.
If you're drowning in newsletters, scattered notes, and generic news feeds, think about what your perfect morning briefing would look like. What would it include? What perspective would it offer? The system is pretty straightforward to build once you know what you want.
]]>When I realized my son was coming, I panicked. How could I keep growing as a person while being there for my child? How could I balance my own goals with being the father I wanted to be?
The first month was rough. There was so much to do, so little sleep, and I felt like I was failing at everything. But slowly, we found a rhythm. I had to let go of my old approach and figure out what actually worked in this new reality.
With a newborn, I managed to build a multi-portfolio investment strategy, develop my AI trading recommendation engine, and work on several other AI automation projects. People ask me how I got all that done. The answer isn't working harder or sleeping less. It's working completely differently.
The biggest adjustment was accepting that I wouldn't have long, uninterrupted stretches of time anymore. I had to learn to work with whatever little moments I could find. I started listening to audiobooks during quiet moments, reading articles on my phone while Flynn slept, watching YouTube videos while doing dishes, taking notes in short spurts throughout the day.
What surprised me was how these little moments added up. Some days I actually learned more than I used to in my long, uninterrupted reading sessions. It's not always perfect, and some days I get nothing done, but I've learned to be okay with that.
When I was backtesting different portfolio strategies, I'd run simulations in Excel while Flynn was sleeping or playing with Inwu. Five minutes here, ten minutes there. Over weeks, those fragments added up to months of solid research work that led to the portfolio system I use today.
Time-blocking stopped working for me. I couldn't say "I'll read from 7-8pm" because that's when Flynn might need me. Instead, I started building situational habits using a simple "when this, then that" approach.
When I wake up, I read for the first 15 minutes. When Flynn goes down for a nap, I open my book. When I take out the trash, I put in my earbuds. When Flynn's asleep for the night, that's when I work on side projects like my AI trading engine or morning briefing system.
It's not perfect, and some days none of these happen. But having these flexible triggers means I don't have to think about when to do things. I just wait for the moment and act.
This was the hardest lesson. I couldn't do everything I used to do. Something had to go.
For me, it was Twitch. I'd spend hours watching game streams, telling myself it was relaxing and helped me focus. But when I tried going three days without it, I realized how much mental space it was taking up. I had room to actually think again.
I still watch sometimes. I'm not perfect. But cutting that one thing out made room for what mattered more. That mental space is where I figured out the holistic portfolio approach, where I designed my AI council of perspectives for the morning briefing, where I solved problems with the trading recommendation engine.
Becoming a father changed everything. I'm not saying I've figured it all out, and most days I'm still figuring it out. But I'm learning that you don't have to choose between being a present parent and growing as a person.
With a newborn, I built systems I'm still using today: a multi-portfolio investment strategy that balances our family's financial future, an AI trading engine that processes information I couldn't possibly analyze manually, a morning briefing system that saves me hours each week. None of that came from long focused work sessions. It came from working with the time I had, building situational habits, and cutting out the noise.
Some days that works great. Other days I watch Flynn play and realize that's the only thing I needed to do today anyway.
For more on what I built during this time, check out Building a Strategic Portfolio System as a New Parent and How I built my own AI council to start every day.
]]>