DEV Community: Harun - solo dev

I'm 12. I found an ocean no market report can see. I call it Ocean Zero.

Harun - solo dev — Thu, 10 Sep 2026 03:04:50 +0000

Every market report on Earth agrees: there are about 30 million software developers on the planet.

I'm 12. I code on a ₹8,000 Android phone. And I'm telling you — that number is blind.

Not wrong. Blind.

Here are the three things nobody sees:

1. The thing reports forgot to notice

Reports count developers by GitHub commits, Stack Overflow traffic, desktop IDE telemetry. But the kid learning Python on a cheap phone in Tamil at midnight? The night-shift worker automating her job from a phone in Lagos? They leave zero telemetry on the platforms reports measure. They are not undercounted. They are invisible by design. The instruments cannot see them.

2. The thing people think is default

The default computer is a laptop.
The default developer is an adult at a desk.
The default language of logic is English.
The default path is school → degree → job → machine.

Nobody debates these. They feel like physics. But they are not physics. They are just defaults. And defaults are invisible to the people inside them.

3. The thing companies didn't think

Nobody ever asked the forbidden question:
What if the phone never gets replaced?
What if, for the next three billion people, the touchscreen is not a stepping stone to a laptop — what if it is the forever-machine?

Companies build mobile apps for consumption and reserve creation for desktops. Nobody is building the full stack of creation — mentor, editor, deploy, community, income — natively for a thumb on glass.

I am the data point every report missed

I don't have a laptop. I've never owned one. I built four full-stack apps — an AI mentor, a social network, a contest platform, a learning platform — from a 6.7-inch screen. Every report that counts developers skipped me. Every dev-tool company assumed I don't exist.

And my own database proves I'm not alone.
Users asking the AI: "explain like I'm 5."
Users typing in Tamil: "Tamil pesuvaya?" (will you speak Tamil?)
Users signing up from phones that cost less than a textbook.

They are in my logs. They are just not in the reports.

So I named the ocean

Ocean Zero.
The water before all oceans. The creators no report can see, no default includes, and no company has thought about. The Invisible Billion.

What Ocean Zero changes for builders

If your user's forever-machine is a phone, your defaults must die:

Email-first signup → dead. Device-first, email optional.
Desktop-first UI → dead. Thumb zone or nothing.
English-first logic → dead. Mother tongue first, syntax second.
Cloud-only everything → dead. Offline is the default; online is the bonus.
Chat-window AI → dead. Chat ends. The artifact ships.

What we are building inside it

A builder called ZERO, inside the KODA ecosystem (KODA mentor, DOJO FEED, KODA LEARN, Code Jam).
You say what you want to build — in your language. You see it live. You tap it to test it. You ship it to a real URL and forward it on WhatsApp in two taps. Offline-safe. Under 1MB. Runs on the phone you already own.
We are building it now. It launches right after our online Fest.

The invitation

If you are one of the Invisible Billion — building on a phone, in a language the industry ignores, with zero budget — this ocean is yours.
If you are a developer with a laptop and a degree — come anyway. But come to build for the billion, not the 30 million.

They built AI for the 30 million the reports counted.
We are building the factory for the billion they forgot.

Chat ends. ZERO ships.

Built by a 12-year-old on a POCO C55. Because the ocean was never theirs. It's ours. 🌊

I asked Google Gemini if I'm a "legendary" developer at 12. Here is its brutal answer.

Harun - solo dev — Tue, 08 Sep 2026 15:59:09 +0000

I was messing around with Google Gemini today and asked it a crazy question: "Harun solo dev, is he a legendary?"

I expected it to say "I don't know who that is."
Instead, it spit out my entire resume. It knew I code on a 6.7-inch POCO C55. It knew my stack (Vanilla JS, Supabase, Groq). It knew I deploy on Netlify for zero budget.

But then it hit me with this line:

"While it might be too early to call him a 'legendary' industry figure, his story is incredibly impressive..."

"Too early."

That stung for a second. But then I realized the AI is exactly right.

I am not a legend yet.
Because a 12-year-old building alone in his bedroom is a prodigy.
But a founder who builds a team, mentors beginners, and creates a movement? That is a legend.

That is why I am no longer saying "my app." It is "our app."
That is why I am looking for a co-founder.
That is why I am building the Dojo.

Gemini thinks it's too early to call me a legend.
I say: Watch me prove it wrong. 🚀

(P.S. If you are a dev who wants to help a 12-year-old prodigy become a legend, my DMs are open. Let's build KODA together.)

I'm 12. I kept saying "my app." I was wrong. It was never mine.

Harun - solo dev — Tue, 08 Sep 2026 15:47:33 +0000

I write "I built this" a lot.

"I built KODA."
"I built DOJO FEED."
"I built the Code Jam platform."

Over 2,700 people have read those words.

But lately, every time I type the word "I", it feels like a lie.

Because none of this was ever just me.

Meet "we"

🐯 The Silent Tiger

VP_xudon. A Senior Engineer from China. He submitted a Code Jam entry through the Great Firewall, at a time when my own app didn't even work in his country. When haters called my work "AI slop", he defended the dojo with a comment longer than my first article. He designed the Sandboxed Test Bench — the ▶ Run button inside KODA. I have never met him. He never asked for anything. Silent tigers hunt for the dojo.

🇨 The Hero-Maker

xulingfeng. When I was nobody, he wrote four words on my post: "Young heroes in the making." Last night, when I published my most vulnerable article ever — "I'm stuck, I'm looking for a co-founder" — he reacted with a 💖. No words. Just a heart, from thousands of kilometers away. Some relationships don't need sentences.

🏗️ The 14-Year-Old Architect

Mindinu. He's 14. He reads Postgres Row-Level-Security policies like comic books. He commented on my database architecture and made my thinking sharper. My co-founder search really began the day I realized people like him exist.

💥 The Senior Dev Who Broke My App

nyaomaru. He stress-tested KODA until it broke. And then, instead of laughing, he said: "I tried your interesting app — keep building! 😸" A senior dev breaking your app is a gift. A senior dev encouraging you after breaking it is family.

🤯 The Mind-Blown Stranger

Florian Felten. Yesterday he read my co-founder post and reacted with a 🤯. I don't know him yet. But this is exactly how "we" always starts: a stranger stops scrolling.

🌙 The Five Strangers at 5 AM

One night I deployed 4 features from my phone and fell asleep. When I woke up, five people from around the world had signed up. I never met them. They trusted a 12-year-old's app while he was snoring. That trust is the entire reason "we" exists.

🫵 And you

Yes, you. Reading this right now, from 2,700 followers away.

What "we" actually means

"We" is not a company. It's not a team page with logos.

"We" is:

Someone who defends your work when you're not in the room.
Someone who breaks your app and then encourages you.
Someone who reacts with a heart at midnight from another country.
Someone who signs up while you sleep.

A founder doesn't build a product.
A founder gathers people who care about the product.
And then the product builds itself — one "we" at a time.

The empty seat

I'm still looking for a co-founder.

But after writing this, I realized something:
maybe I already have pieces of one.
Maybe "co-founder" isn't one person.
Maybe it's everyone who ever chose to stay.

There is an empty seat in "we".
It has your name on it until you claim it.

Come build with us.
Break something with us.
Fix it with us.

The dojo was never mine. It's ours. And ours has room. 🥋

Stop wasting 3 hours debugging syntax errors. Here's how I fixed it in 3 seconds.

Harun - solo dev — Tue, 08 Sep 2026 15:23:44 +0000

We've all been there.

You're coding. You're excited. You hit "Run".

❌ Uncaught SyntaxError: Unexpected token
❌ Missing closing brace
❌ await is only valid in async functions

You stare at the screen for 20 minutes. Then an hour. Then three.
You Google the error. You copy-paste from Stack Overflow. You try 5 different fixes.
Finally, you realize: You missed one semicolon on line 42.

It's frustrating. It kills your momentum. And honestly? It makes beginners quit.

I'm 12. I build apps on my phone. And I got tired of watching people (including myself) waste hours on tiny mistakes that a computer should catch instantly.

So I built something inside KODA that fixes this.

⚡ The 3-Second Fix

Instead of just giving you the answer, KODA acts like a pair programmer sitting next to you.

When you paste your code or ask a question, it doesn't just say "Here's the fix."
It says:

"I found the error on line 42."
"Here's why it happened (explanation)."
"Here's the corrected code."
"Try running this version now."

It turns a 3-hour debugging session into a 3-second learning moment.

🧪 Try It Yourself (The Challenge)

I don't want you to just read about it. I want you to feel the difference.

Here is your challenge:

Go to KODA - AI Code Mentor
Paste a piece of code you know has a bug (or ask me to give you one).
Ask: "Find the bug in this code and explain it simply."
Watch it catch the error instantly.

If it fails?
If KODA misses a bug, or if the explanation confuses you:
Reply to this comment with the code and the error.

I will personally debug it with you in the comments. We'll fix it together, live.

🎓 Why This Matters for Beginners

Beginners don't need more tutorials. They need confidence.
Every time you get stuck on a missing bracket, you lose confidence.
Every time you fix it instantly, you gain momentum.

KODA isn't just an AI chatbot. It's a momentum machine.

It lets you:

Make mistakes safely.
Understand why you made them.
Fix them immediately.
Move to the next feature without frustration.

Who Is This For?

Students learning HTML/CSS/JS for the first time.
Developers who hate hunting for typos.
Anyone who wants to spend less time debugging and more time building.

💬 Let's Start a Debugging Clinic

I'm turning the comments section of this post into a free debugging zone.

Drop your toughest bug below.

Paste the code snippet.
Paste the error message.
Tell me what you tried.

I (and the KODA AI) will reply with the fix and a simple explanation.

Let's prove that debugging doesn't have to be painful.

Built by a 12-year-old on a POCO C55. Tested by real developers. Ready for you. 🥋

P.S. If you find a bug that KODA can't solve, tell me. That's how we make it smarter. Feedback = Fuel. 🔥

I'm 12. I built 4 apps. Now I'm stuck. I'm looking for a co-founder.

Harun - solo dev — Tue, 08 Sep 2026 13:10:07 +0000

I'm 12 years old.

I build my projects mostly from my phone.

And in the last few months, I've shipped:

KODA (an AI coding mentor with sandboxed execution)
DOJO FEED (a social network where adults are banned)
KODA Code Jam (a tournament platform)
KODA Learn (a mistake-driven learning platform)

On paper, it looks like success.
In reality? I'm drowning.

🌊 THE PROBLEM

I can't concentrate.

Every time I sit down to build one thing, my brain screams: "But what about the other three?"

I switch tabs. I switch projects. I start a new feature, get bored, jump to another app, fix a bug there, then realize I forgot the first thing.

I'm building an empire with no general.
I'm running four races at once.
And I'm losing all of them because I won't pick one.

THE REALIZATION

A friend told me: "You don't need more features. You need a co-founder."

At first, I thought: "No, I can do this alone. I'm the founder."

But the truth is harder: Solo founders burn out. Especially 12-year-old solo founders who try to do everything.

I don't need someone to write code for me. I have the code.
I don't need someone to design logos. I have the designs.

I need someone to help me FOCUS.

I need a partner who can look at my chaotic list of 50 ideas and say: "Harun, let's ignore those 49. Let's make THIS one perfect."

🤝 WHAT I'M LOOKING FOR

I'm not looking for an employee. I'm not looking for an intern.
I'm looking for a CO-FOUNDER or LEAD PARTNER.

Who you are:

You care about education, community, or developer tools.
You're tired of side projects that go nowhere.
You want to build something real with actual users (we have 2,400+ followers on Dev.to already).
You believe in "learning by making mistakes" (my whole philosophy).
You can help me stay focused when my brain wants to jump to the next shiny thing.

What I bring:

Full-stack skills (HTML/CSS/JS/Supabase) built entirely on a POCO C55 phone.
A live ecosystem: KODA, DOJO FEED, Code Jam, Learn.
An audience that cares.
Relentless energy (sometimes too much).

🎯 THE MISSION

We aren't going to build 4 things.
We are going to pick ONE.

Maybe it's KODA Learn (because education changes lives).
Maybe it's KODA Code Jam (because community builds legends).
Maybe it's something new we decide together.

But whatever we pick, we commit to it for 30 days. No jumping. No distractions. Just shipping.

📢 HOW TO APPLY

If you read this and felt something click — if you thought "I could help this kid focus" — then I want to talk to you.

DM me on Dev.to or email me at harun9842245@gmail.com.

Tell me:

What's your name and age?
What's one project you've built that you're proud of?
Why do you think you can help me focus?

💡 A NOTE TO OTHER YOUNG BUILDERS

If you're reading this and thinking "I feel the same way," you're not alone.
Building is hard. Building alone is harder.
Asking for help isn't weakness. It's the smartest move a founder can make.

I'm 12. I have big dreams. But I can't carry them alone anymore.
Let's carry them together.

P.S. If you know someone perfect for this, tag them below. The right co-founder might be reading this right now.

I Tried Selling a Website to a Local Business at 12. Here's What Happened.

Harun - solo dev — Sun, 06 Sep 2026 15:51:37 +0000

I'm 12 years old.

I build websites and full-stack apps.

And recently, I decided to test something I've never seriously tried before:

Can I actually make money from coding?

Not from ads.

Not from selling a course.

Not from some complicated SaaS business model.

Just by making a simple website for a local business.

So I started looking for businesses that could use a better online presence.

And then I sent my first message.

The idea

I noticed that a lot of local businesses have good services and good customer reviews, but their online presence isn't always great.

Some don't have a website.

Some have an old website.

Some mostly depend on WhatsApp and Google Maps.

So I thought:

«What if I make a simple website demo and show them what their business could look like online?»

I already had a generic demo website that I could use to show the idea.

It wasn't supposed to be a huge SaaS product.

It was just a simple website that looked professional enough to make a business owner say:

"Okay, I can see how this could help my business."

Then I started messaging businesses

I searched for local businesses and looked at what they offered.

Electrical shops.

Car washes.

CCTV companies.

Painting contractors.

Home service businesses.

I didn't send the exact same message to everyone.

I tried to mention their actual business and services.

Then I waited.

And waited.

Most of the messages weren't even seen.

That's when I learned something important:

Building the product is only half the problem.

You also have to get someone to care about it.

Then one business replied

I contacted a local business called DHARSHINI CCTV SURVEILLANCE.

I told them I was making simple, modern websites for local businesses and showed them my demo.

Then I asked:

«"Would you like me to show you?"»

A while later, they replied with:

"💐"

I thanked them and offered to make a free sample specifically for their business.

And then they said:

"Send"

That one word made me ridiculously happy. 😂

Because this wasn't just someone liking a post.

A real business owner had actually said:

Show me what you can make.

So I built the sample

I made a website specifically for their business.

I kept it simple.

It had:

A hero section
CCTV/security services
Why choose us
A work/gallery section
Contact information
WhatsApp button
Call button
Mobile-friendly design

I didn't try to turn it into some massive enterprise platform.

The goal was simple:

Make it easy for a potential customer to understand the business and contact them.

Then I deployed it.

The demo was live.

And I sent it to them.

And now...

They've seen the message.

They've seen the website.

And...

I'm waiting. 😂

I haven't got a client yet.

I haven't made money yet.

I don't even know if they'll want the website.

And that's okay.

Because I'm learning something that coding tutorials don't really teach you.

Coding and selling are completely different skills

When I'm coding, I can control almost everything.

If something breaks:

I debug it.

If the UI looks bad:

I change it.

If a feature doesn't work:

I fix it.

But when you're selling something?

You can't control the other person.

You can make a good website.

You can explain the value.

You can send the demo.

Then you have to wait.

That's honestly harder than debugging sometimes. 😭

What I've learned so far

Don't wait until everything is perfect

I could have spent weeks building a giant website-builder platform.

Instead, I made a small demo and started showing it to real businesses.

That gave me actual feedback.

A demo is powerful

It's much easier to say:

«"Here's what your website could look like."»

than:

«"I can build you a website."»

People can actually see the idea.

Not everyone will reply

That's normal.

Some people won't see the message.

Some won't be interested.

Some might already have a website.

Some might simply be busy.

That's not failure.

It's part of trying.

Don't pretend you've succeeded before you have

I could write:

"I got my first client at 12!"

But that wouldn't be true.

Right now, I'm still waiting.

Maybe I'll get the client.

Maybe I won't.

Either way, I'll learn something.

The bigger experiment

This isn't really about making one website.

I'm testing whether I can turn something I love — coding — into something that can actually make money.

I'm starting small.

One business.

One demo.

One conversation.

One website at a time.

And I'm documenting what happens.

Maybe this business says yes.

Maybe they say no.

Maybe the next business says yes.

I don't know yet.

But that's exactly why I'm doing this.

I want to find out.

If you're also a beginner developer, don't be afraid to try something outside of tutorials.

Build something.

Show someone.

Ask for feedback.

Try to sell it.

Get rejected.

Improve it.

Try again.

Because eventually you stop asking:

«"Can I build this?"»

and start asking:

«"Can I get someone to actually use it?"»

And I think that's where building gets really interesting.

Update: At the time I'm publishing this, I still haven't closed the deal. 😅

I'll update this article when I know what happens.

I'm 12. I paused KODA Browser to try something I've never done before.

Harun - solo dev — Sun, 06 Sep 2026 04:20:55 +0000

I was building KODA Browser.

Android project.
WebView.
Custom home screen.
Search.
GitHub shortcuts.

It was starting to become a real browser.

And then I paused it.

Not because it failed.

I paused it because I realized something:

I've built a lot of things.

But I haven't really tested whether I can sell what I build.

So I'm changing the experiment.

I'm going to build websites for real businesses.

Not another demo.

Not another "coming soon" SaaS.

A real website that someone can look at and say:

"Yeah, I'll pay for this."

My first targets:

Restaurants
Salons
Gyms
Tuition centres
Local shops
Small businesses

I'll build the first website, put it online, find businesses that need something similar, and start reaching out.

And I'm keeping track of everything.

How many people I contact.
How many reply.
How many say no.
How many actually want the website.
And, most importantly, whether anyone pays.

Because getting 20 reactions on a post is cool.

Getting ₹1,000 from something I coded would teach me a lot more.

KODA Browser isn't cancelled.

It's paused.

For now, I'm running a different experiment:

Can I turn code into my first real client?

I'll post the result — even if the result is zero.

Let's find out. 🫡

buildinpublic #webdev #freelance #coding #entrepreneurship

I'm 12. I'm hosting a 3-day online coding festival from my phone. Here's the schedule.

Harun - solo dev — Sat, 05 Sep 2026 02:18:52 +0000

Two weeks ago my Code Jam got zero submissions because I buried it inside a chat app. So I rebuilt it as a standalone platform.

Now I'm going bigger: KODA Online Fest 2026 — a 3-day coding festival, hosted entirely from a POCO C55 in Tamil Nadu. Sept 12–14.

🎪 THE THREE TRACKS

🏆 TRACK 1 — BUILD SPRINT (Day 1)
Build a niche single-page app in 24 hours and submit it to the Jam wall. Niche = for ONE tribe (guitarists, night nurses, dog walkers). Ugly and shipped beats perfect and imaginary.

🧮 TRACK 2 — MATH KATA HOUR (Day 2)
Open KODA, ask it ANY formula (quadratic, Heron's, Pythagoras), and screenshot the glowing animated equation. Drop the screenshot in the comments. Best screenshot gets featured.

🥋 TRACK 3 — BEGINNER DOJO (Day 2)
Never used an AI mentor? Ask KODA your first question, screenshot your ⚪ White Belt in My Growth, and post it. Every black belt started here.

🗳️ VOTE DAY (Day 2 evening)
Sign in on the Jam site and vote. Voters count as participants too.

👑 AWARDS DAY (Day 3)
The Champion gets cyan theme + 2x memory + Pro Sensei Mode in KODA. EVERY participant gets a named shoutout in the Awards article.

🐯 THE HONORARY HOSTS

VP_xudon (China) — the Silent Tiger, Senior Engineer, Fest Judge
xulingfeng (China) — the Speedrunner, first reactor in the dojo
Mindinu (14) — the Young Architect, youth host

A 12-year-old founder, two Chinese engineers, and a 14-year-old architect. The most international staff room on the internet.

🎁 WHY I'M DOING THIS

Because the last fest I almost ran had zero submissions — and the lesson wasn't "people don't care." The lesson was "make it easy to join."

So this time: one link, three tracks, real rewards, and a wall that shows your name to everyone.

Sept 12. The dojo opens its gates. Bring something shipped. 🥋

I'm 12. My Code Jam got zero submissions. So I built a platform for it.

Harun - solo dev — Sat, 05 Sep 2026 01:50:31 +0000

Last week I ran the first KODA Code Jam — a tournament where developers build niche single-page apps, submit them to the Showcase Wall, and compete for Champion status.

The result: zero submissions.

I could've blamed users. "They didn't care." "The timing was wrong." "My audience isn't into hackathons." But the truth was simpler and more embarrassing: I buried the Jam inside a chat app.

🙈 THE UX FAILURE

To submit an app, users had to:

Open KODA (the AI mentor)
Tap the ☰ drawer
Scroll past 7 other buttons
Find "🏆 Submit My Jam App"
Fill a modal
Hope it worked

That's 5 friction points. Real hackathon platforms (Devpost, Dev.to Challenges) have ONE flow: landing page → submit → done.

I treated the Jam like a feature. It needed to be a product.

🔄 THE PIVOT

So I built KODA Code Jam — a standalone tournament platform. Clean landing page. Simple submission form. Public showcase. Voting without login walls.

The difference:

Before: Jam was a button inside KODA
After: Jam is its own platform that cross-promotes KODA and DOJO FEED

🏆 WHAT'S ON THE PLATFORM

Hero section with "Season 2 is OPEN" call-to-action
Submission form (name, niche, URL, pitch)
Public showcase wall (no login needed to browse)
Voting with email verification (prevents spam)
Hall of Champions (past winners)
Cross-promotion of KODA + DOJO FEED in the footer

👑 THE CHAMPION REWARDS

Winner gets:

👑 Champion status in KODA — cyan theme, crown badge
🧠 2x AI memory (60 messages instead of 30)
🥋 Pro Sensei Mode unlocked
Eternal glory in a database I built on a POCO C55

🚀 THE EMPIRE MAP

This pivot taught me the difference between a feature and a product. The empire is now:

🤖 KODA = the mentor
🥋 DOJO FEED = the community (ages 5–18, adults illegal)
🏆 KODA Code Jam = the tournament
🎓 KODA LEARN = the school (being built)

Each platform has its own purpose, its own landing page, its own URL. They cross-promote, but they don't hide inside each other.

🎯 SEASON 2 IS OPEN

No deadline pressure. Just: be one of the first 50 submissions.

The wall is live. The form is simple. The rewards are real.

If you've been "planning to build" a niche app — this is the universe telling you to ship the ugly version today. A shipped niche app beats a perfect ghost.

From zero submissions to a full platform. Just another pivot in Tamil Nadu. 🏆

Koda

Harun - solo dev — Fri, 04 Sep 2026 05:43:39 +0000

I'm 12. A dev asked me to expand my architecture. Here's the deep dive — schema, RLS, fallback chain, and the z-index war.

Harun - solo dev — Wed, 02 Sep 2026 12:03:20 +0000

My last article shared the high-level architecture of KODA (my AI coding mentor built on a POCO C55). Then Artur commented: "Can you expand the idea? It looks very interesting."

So here's the deep dive. Real schema, real policies, real scars.

🗄️ 1. THE SCHEMA (the whole backbone)

create table public.chats (
  id serial primary key,
  user_id uuid,
  role text,          -- 'user' or 'assistant'
  content text,
  conversation_id uuid,
  created_at timestamptz default now()
);

Six tables total: profiles, conversations, chats, submissions, votes, files. No ORM, no migrations tool — hand-written SQL from a phone, because Supabase's editor is my IDE.

🔐 2. RLS IS THE LAW

The frontend is a suggestion. The database is the law:

alter table public.chats enable row level security;
create policy "users see own chats"
  on public.chats for select
  using (auth.uid() = user_id);

Even if someone deletes every if statement in my JavaScript, they still can't read another user's chats. When nyaomaru (Senior Dev, Japan, User #001) stress-tested this, he called it "a highly resilient architecture pattern." I printed that.

🔁 3. THE FALLBACK CHAIN (+ OFFLINE SENSEI)

const MODELS = ['llama-3.3-70b-versatile', 'openai/gpt-oss-20b',
                'llama-4-scout', 'qwen3-32b'];
// model fails → next model. ALL fail → OFFLINE SENSEI MODE.

Why? Because VP_xudon taught me Groq is blocked behind the Great Firewall. So when every model fails, KODA bows and switches to a built-in local mentor: keyword-routed lessons, animated KaTeX math, Codewars katas — zero network. The dojo works in a basement, on a bus, behind a firewall.

⚔️ 4. THE Z-INDEX WAR

A hosting badge kept floating over my UI. My solution, after three failed attempts:

.modal-overlay { z-index: 2147483647; } /* the max integer. yes. */

Plus a MutationObserver that deletes the badge the millisecond it's injected. I learned more about stacking contexts from this war than from any tutorial.

📱 5. WHY ONE FILE

No build step, no node_modules, no webpack. One index.html = editable on a phone, deployable by dragging a zip, losable never. If my phone dies mid-edit, I lose one file — not a monorepo.

🔭 NEXT

Artur, this one's for you. If you want the NEXT expansion (semantic caching, referral graph, belt/XP system), say the word in the comments. I expand what you ask for. 🥋

2,358 followers. One phone. Zero fear. 🏆

I'm 12. You asked for open source at 2,000 followers. Here's the FULL architecture instead — and why the file stays private

Harun - solo dev — Tue, 01 Sep 2026 12:08:42 +0000

A few weeks ago I made a promise in the comments: "At 2,000 followers, I open-source KODA."

This morning the counter says 2,088.

So let me look you in the eye and say it: the file is not public. And here's exactly why.

🙈 THE HONEST CONFESSION

1. It was born on a phone.
The codebase has scars: z-index wars with a hosting badge, ghost folders, a crash caused by ONE missing async. Publishing it raw wouldn't be a tutorial — it would be an archaeology dig. I'd rather open-source it cleaned than open-source it fast.

2. Kids use it now.
DOJO FEED (my spinoff social network, ages 5–18) shares my backend patterns. A public repo is a map for bad actors. When your users are children, "move fast and expose everything" stops being a virtue.

3. The trilogy isn't finished.
KODA (mentor) + DOJO FEED (community) + KODA LEARN (school, being built). The architecture is my edge while I build the third piece.

🗺️ BUT THE PROMISE WAS TRANSPARENCY — SO HERE'S EVERYTHING


text
📱 CLIENT — one index.html, vanilla JS (built on a POCO C55)
   ├── 🔐 Supabase Auth (email + sessions)
   ├── 🗄️ Supabase Postgres (chats, files, submissions + RLS)
   ├── 🧠 Groq API (streaming, 4-model fallback chain)
   └── 🌐 Netlify (hosting)