I composed this poem in my head last Saturday around 4 am, when I was rocking my little one to sleep.

A Father's Fate

Dear God, 
First I pray you protect, bless, and guide
My wife and my child. 
I pray you do the same for me too, 
But do it third. 
And if you should bless us with another, 
Then down the list I go - 
And so on and so forth. 
Love, health, joy, strength, and abundance - 
May these five be the pillars that hold us. 
May none within this family see six feet before me. 
But long before then,
May many decades pass with these loves as my seventh heaven. 
Dear God, I pray this be my fate.

I.I.

P.S.: You can listen to me reciting it if you press the ‘Play‘ button at the top of this post if you haven’t done so already. And let me know what you think in the comments.

If you’re new here and would like to learn more about me and what Augmented Ivan is…

Start Here

This one is a favorite and reminds me of why I write poetry. I wrote it in 2015, and I’m still looking out for Later (a future version of me that I created while writing this poem) — waiting to meet him again. I think he’d be happy with the current version of me, but he’d probably wish I wrote more nowadays.

The videos I recorded during that time really helped me maintain my sanity. I’d try to record at least one every week and share them on weekends.

Also, would you just look at that hair? Majestic! Man, I miss this guy.

Fun fact: the hair in this and other videos is an indicator of how long the lockdowns lasted because I wasn’t getting my hair cut during that time.

Also also, forgive the video and audio quality. I did the best with what I had at the time, which was just a smartphone.

I’ll be sharing more recordings in the coming days. Hope you enjoy them and get a better understanding of what my poetry feels like.

I would love to hear your thoughts on the poem and the idea of creating to be remembered. Please let me know in the comments.

Ivan

“Large language models (LLMs) often ‘satisfice’ — they accept good-enough rather than optimize.”

That’s what Claude replied when I asked it about creating a self-scoring skill after a colleague shared a tip that involves asking your LLM to score its output and following that up with something like, “What would you change to improve your score?”

I looked it up, and satisficing is a blend of “satisfy” and “suffice.” It describes the tendency to settle for a solution that meets the minimum threshold rather than push toward the best possible one. Turns out, it’s not just a human thing. When an LLM finishes a task, it tends to stop at it works instead of it’s good.

That gap — between working and good — is where a lot of value quietly leaks out, and you might not know it unless you ask.

So we built a quality gate

Claude and I built claude-self-score, a Claude Code plugin that adds a structured quality gate before task completion.

The idea is simple: instead of letting Claude declare “Done!” the moment a task is technically completed, the plugin asks the LLM to evaluate its own work against a defined rubric — and only close the loop if the output actually meets the bar.

No more bare-minimum output.

How it works

Claude scores its own work across 7 dimensions:

1. Correctness — Does it actually do what it’s supposed to do?

2. Code quality — Is it clean, readable, maintainable?

3. Completeness — Are all requirements addressed?

4. Elegance — Is it the simplest solution that works?

5. Performance — Does it use resources wisely?

6. Security — Are there obvious vulnerabilities?

7. User intent alignment — Does it meet the literal request and the actual intent behind it?

Each dimension is scored on a 1.0–10.0 scale. If the overall average falls below 9.5/10, the plugin identifies specific improvements and asks before implementing them. It also detects dishonest self-scoring patterns — things like score inflation and N/A abuse, where a model might quietly sidestep dimensions it’s underperforming on.

And it runs automatically at the end of every task. No need to manually invoke it.

Why this matters

Without a quality gate, AI assistants do what humans do under time pressure: they stop when something works, not when it’s right.

Structured self-reflection catches things that a passing glance misses — edge cases that weren’t obvious at first, code that functions but would be a headache to maintain, and solutions that answer the question asked while missing the point. That last one is subtle and common. An LLM can satisfy every stated requirement and still produce something that doesn’t serve what you actually need.

The improvement loop also has a compounding effect. Small, targeted fixes — the kind that only surface when you’re forced to look critically — add up.

Oh, and it failed its own bar at first

Crazy thing is, since it automatically runs at the end of each task, the skill ran after its creation — self-scoring and improving itself. The first pass came in at 9.4 — just below its own 9.5 threshold.

Basically, the skill designed to catch “good enough” caught itself being good enough.

There’s something humbling and deeply satisfying about that. The system worked exactly as intended, on the thing that built it.

Try it out, and let me know how it works for you.

Get The Plugin

Have you ever noticed your favorite LLM just satisficing it? Let’s talk about it in the comments.

AI use in this post:

• Banner image was created with Claude and Nano Banana 2.

Funny thing is, I found that relationship after I first performed it at my EP launch in 2018. You can learn more about that here.

EP available here.

Track produced by CmoneY
Lyric video by Kelvin Kaesa

And what better way to do that than to tell you about the best poem I’ve ever written and recorded — at least to me.

What is REAL?

REAL is the title track from the extended play (EP) project of the same name. My friend CmoneY produced the entire project, and another friend, Kaesa, created the above lyric video. This was way back in 2018, when I was a 27-year-old trying to figure things out — to be fair, I’m still trying to figure things out, even if I’ve figured quite a few things out in the past 8 years. Lots of figuring out, such is life.

The story behind REAL

I can’t remember the exact day when I wrote the poem (I should really date my pieces), but I think it was the year before the project. What I do remember quite vividly is that I wrote it after an Nth breakup — of course I did, everyone knows an artist’s best works require a dose of character development.

Also, I know what you’re thinking, “Nth?” What can I say? A guy was trying to find love in the hard dating streets of Nairobi. Roughest dating streets on Earth. You find love here; you can find it anywhere.

The point is, I wrote REAL because I was tired. Tired of trying and failing. Tired of putting my all into something and ending up with nothing. Tired of not having something... real. So I wrote about what I wanted in a relationship.

Side note: a few years before this, in 2015, a poet I admire, Bassey Ikpi, told me, during a workshop she’d invited me to, that I should write about my own experiences — that I should write something... real. Back then, I was a young poet who sometimes wrote poems that sounded good. The title REAL might have been inspired by this interaction as well.

Why is REAL my favorite poem?

Not only is REAL my favorite poem because of how honest it is, but it’s also my favorite because I consider it the foundation of everything I’m blessed to call mine today.

See, on the night of the EP launch, when I first performed REAL, there was a certain lady in the audience. I’d known her since uni days, when we were acquaintances and neighbors. She’d heard that I was launching an EP and was looking for something interesting to do that weekend, so she attended.

According to her, she loved the poem and would listen to it often in the days that followed. I do have that effect on people (pats self on back). Thing is, she too was tired of failed relationships — I’m not even kidding, the dating streets of Nairobi are like Character Development Premium Pro Max, ask anyone who lives here.

A few months later, she and I started talking on Facebook. We mostly talked about our favorite show, This Is Us (this would go on to inspire the title of my first album, Us, released in 2019, before the pandemonium).

Anyway, I asked her if she’d like to get coffee sometime, and that date led to another. We started dating and days turned into months; months turned into a year; a year turned into seven and counting at the time of writing. She and I have built a life together, gotten married, and are still building a family. Till this day, she argues that the coffee date wasn’t actually a date. I ask her to look at her left hand for proof that it was indeed a date.

Other side note: she still loves the poem so much that she asked me to include quotes from it on a page, and she framed it.

But you see what I mean by REAL being the foundation, right? Everything I talk about in the poem is what I experience today with her. The most honest poem I ever wrote became the truest thing I’ve ever lived.

REAL is everything I have; everything I have is... real.

Listen to my projects

If you love the song REAL even half as much as I do, I highly recommend (self plug) listening to the EP and some of my other projects, including:

• Us — A live performance album recorded with an incredible audience back in 2019! Us is part interactive storytelling, where I let the audience decide what goes onto the album, and part true story of the early days of my relationship with my then girlfriend - now wife.

• Un Jour (One Day) — A piece recorded in both French and English with the same message: Know that I lived.

Listen to all of my projects here.

Ivan

If you like my writing, poetry, and music, consider subscribing to see what else I get up to and create. I’m on a journey to create again after a long hiatus, and I’m documenting my authentic exploration of art, AI, and self along the way.

Subscribe now

Ever since OpenClaw became a thing, with every tech personality/influencer hyping it, I’ve been thinking about how to replicate what it does.

Why? Simple. I can handle the setup process it requires, and I don’t even mind setting up the security measures — it’s a learning process, and I enjoy learning new things. But the one thing I don’t like is…

Using the Claude API with OpenClaw gets expensive

See, I have a clear preference when it comes to LLMs: Claude. I’ve preferred using Claude for some time now — long before recent events. My friends and colleagues are probably tired of hearing me talk about it for the past year or so.

So when I tried out OpenClaw on a VPS for the first time, I naturally selected Claude as my OpenClaw’s brain.

The problem is that I couldn’t use my Claude Max subscription — only the Claude API — and I wasn’t going to risk getting banned by Anthropic. And that got too expensive for my blood, quick. Setting things up cost me around $30, even though the first rule I put in place was smart model selection (where certain models are used for specific kinds of tasks to minimize costs). None of what I did required intense coding.

Doing the math led me to one clear conclusion: this setup is pricy.

Also, I don’t like uncertainty when it comes to money.

But wait, there’s an alternative

Last week, Claude Code introduced a Remote Control feature — enabling users to run Claude Code on a phone, tablet, or browser from anywhere as long as the terminal remains active.

That got me thinking: How can I replicate OpenClaw using this feature?

I recently learned how to set up a VPS so that I could use OpenClaw, and most of its features are just based on .md files and cron jobs.

So I spent a few days trying to build my own 24/7 assistant that uses my Claude Max subscription.

Then I documented the process using Claude and the new Gamma connector — Claude really is coming through with the best features lately.

Here it is for you to download and build your own! Enjoy!

Build Your Own Always On Claude Code Assistant (v2)

815KB ∙ PDF file

Download

A guide to running Claude Code on a VPS with Heartbeat, cron jobs & Remote Control

Download

It’s a bit more technical than I would’ve wanted it to be — primarily because of the steps needed to keep everything alive — but I think it’s a decent alternative to the Claw.

Please let me know if you have any questions or if you need assistance with anything.

AI use in this post:

• Banner image created with Canva AI.

• Slide deck created with Claude and Gamma.

Edit: I forgot to mention in a previous version of the slide deck that you should select Ubuntu 24.04 LTS as your VPS’s operating system. This has been added in the latest version of the deck.

I’m not sure how to start this. How do I take the first step on a journey that I find… daunting? I’ll be honest: I’ve hesitated starting this Substack for months. But like with most things, I guess the best way to start is to say, “F*** it,” and just do it.

So here I am. Hi, my name is Ivan, and I’m a poet, storyteller, communications lead, and AI enthusiast trying to figure out and record who I am in this age of augmentation. Because, ultimately, that’s what creative self-expression is: introspection.

But that doesn’t tell you much about me, so let’s just dive into some of the questions you might have.

Who am I?

I’ve often heard that what one does for a living doesn’t really tell you much about the person and that they’re more than their titles (this is maybe more significant today in this AI era). You’re supposed to give a deeper, more meaningful explanation of who you are.

So here are a few facts about me:

• I love “Love.” That’s what my wife tells me. I love the idea of it, love listening to people speak about it, love watching people express it, whether it’s romantic, platonic, familial, or love of self. I write about it a lot in my poetry - and I also write about life’s lessons and the pursuit of dreams.

• I’m a recovering perfectionist overthinker who, for a while, managed to just put poems – and myself – out there without obsessing over them. It helped me become proficient at publishing my work, but I now find myself having to try to shake off the overthinking again.

• I’m also a planner who likes to break things down into processes and figure out how to improve them. I’d say that’s where my enthusiasm for AI and agentic workflows starts. I see AI as a tool that can save me time in my professional (I like to work smart, not hard) and personal life. It also allows me to bring ideas to life that were stuck in my head.

• My wife will tell you that I tend to see the world in black and white, but I’m really trying to add a healthy amount of grey to my perspective. I promise I am. Take AI, for example: some hate it and don’t want to use it, and others think it should be in everything. I certainly don’t think it should be everywhere (especially on apps suggesting what to write to a friend or loved one, ugh), but it does have its use cases (mostly for work within organizations).

• Like many artists, I’m an introvert who protects his space and – more importantly – his peace. Yet, for some reason, we expressive introverts willingly step on stages to lay our souls bare for all to see. We’re a crazy bunch. Life seemed simpler when we’d create alone, in our rooms. Now we’re just… out here, outside.

• Lately, I’ve been admiring forests – not walking in them, I’d easily get lost, and my relationship with gluten means I likely wouldn’t have enough breadcrumbs to trace my way back where I live. No, I mean staring at them and getting lost in their beauty (from afar). I believe it’s a consequence of spending most of my time on screens and in the digital space. I now crave nature and what’s real – the sounds and sights of trees moving gently with the wind; the songs birds sing; the colors in the sky at sunset and sunrise. I see a forest, I stop and stare.

• I’m a big proponent of artistic self-expression – it allows you to discover yourself and gain a deeper understanding of the world around you. In fact, I once gave a TEDx talk on it, which you can watch below.

You’ll learn more about me through my work – maybe you’ll know more about me than I do. I believe that there’s more of me hidden in between the lines than in the words I write.

What is Augmented Ivan?

Augmented Ivan is my digital garden – the space where I express myself and where you discover the duality within me:

• The perfectionist expressing his flaws

• The man who seeks natural beauty but gazes at the digital world

• The planner trying to go with the flow

• The human trying to master his demons

• The curious introvert sharing what he knows (and what he doesn’t)

• The artist and the employee

This space is meant to hold everything I feel comfortable discovering and sharing with the world. It’s also meant to record how I, as an artist and human, evolve – hence the “Augmented” bit.

Looking back on the past 12 years, I’ve changed significantly through different phases. I’ve gone from amateur poet to spoken word artist, then poetry event organizer, then interactive storyteller, and even comedy night organizer for a bit. Then the pandemic hit, and things simply stopped. I stopped creating for the love of it, became more withdrawn than I was before, and focused on creating content for paychecks.

You and I now find ourselves here: the phase in which I feel the need to create and express myself again. I don’t know if I can be the talented poet and storyteller I once was, but I’m looking for him.

The world that part of me thrived in has changed, however. We now live in a time when (almost) anything is possible. Where ideas are a few prompts away from ‘existing’. So there’s also that to figure out: how do I best bring my ideas to life?

Welcome to the Augmented Ivan phase of my life.

Why am I doing this?

Simple. For the same reason I’ve always written and expressed myself: I want to be remembered; to tell the world I was here; to say, “This is who I am.”

I want to attain immortality. Not physically – I don’t think I’d enjoy that, and there’s beauty in the finite – but through my words.

What should you expect from this Substack?

Ever seen a botanical garden where everything is perfectly structured into geometrical areas, and all the flora is cut according to specific measurements? I haven’t, and I wouldn’t want to.

This garden will be more ‘curated chaos’. Creativity, after all, requires a bit of chaos to come to life. So here’s what you’ll find:

• My written poems (old and new ones)

• AI-augmented videos for some of my favorite poems

• Explainer videos on how I use AI

• Thoughts and essays on topics that matter to me or that I’m curious about

• Anything else I think of down the line

I want to make this space as authentic as possible, so here are some ground rules I’ve set for myself:

1. The words are mine: Whether written or spoken, what you read and hear will come from me – not LLMs. I’m no AI or tech expert (yet), and I do not pretend to know it all, but one thing I do pride myself on is my writing ability. My pen game has made crowds laugh, cry, and feel deeply. I cannot leave that to some software.

2. Disclosure of AI usage: You will always find a list of the AI tools I use at the end of each post.

3. No content for the sake of content: I am not an AI influencer, nor can I be one while maintaining my sanity. This tech niche moves at a speed I cannot keep up with. I will not update you on the latest models or tools – others can do that, and I appreciate them for doing so. I’d like to share how I use some of them, but I don’t have the time to post here according to a content calendar. I might go quiet sometimes. Bear with me, I have a family, work, and life in general to prioritize.

How do we start this journey together?

Start by subscribing to this Substack if you like what you’ve read so far. And leave a comment if you’d like to start a conversation, or a “Hi!” if you’d like to let me know you read this entire post.

Subscribe now

I’m truly grateful for the time you’ve given me and my writing, and I hope to see you around!

AI use in this post:

• Banner image created with a prompt from Claude, used in Canva AI.