You only pay if you want to keep it.

I build the AI feature into your product, you watch it run on your own data, and then you decide. If it is not what you wanted, I take the code back, you pay nothing, and you walk. No questions.

1 You see it running first
2 Not right? The code comes back
3 You pay nothing and walk

AI already shipped in two client apps that are live in the stores. My name is on neither listing. Judge the work, not the pitch.

zaiban@greeencode.com How the deal works

One build at a time. Tell me the product and the feature, and you will know quickly whether I am taking it.

2client apps live in the stores
5AI features I built, running in them
iOS + AndroidReact Native, shipped

AlmightyClient app, App Store and Google Play

EzeeNotesClient app, Google Play

The deal

How you end up paying nothing

Every developer you talk to sounds equally confident over email, so I take that part out of it. You do not have to trust me, you have to look at a working build. The risk sits on my side of the table until you have seen it.

01
We agree what working means
Before anything is written down in code, we write down the one thing the feature has to do and how we will both know it does it. One paragraph, not a specification document.
02
I build it and you watch it run
Not a slide, not a recorded demo, not a prototype in a sandbox. The feature running inside your product, on your data, in front of you.
03
You decide, and one option costs you nothing
If it is not what you wanted, I take the code back, you pay nothing, and you walk. No questions, no invoice, no argument about scope. If it is what you wanted, it is yours and we talk about the rest.

I can offer that because I have shipped this before, and because I only run one build at a time. Both of those are checkable: the shipping is above, and the second is why I ask what a project is before I say yes.

Almighty chat screen with message bubbles and a voice note button

EzeeNotes home screen with subject cards for physics, chemistry and maths

Almighty voice call screen showing a connected caller

EzeeNotes screen showing a chapter list inside a subject

Almighty group call screen with participant controls

EzeeNotes screen showing practice questions

Almighty community screen with a post composer and threaded replies

EzeeNotes notes page with a QR code that opens a video explanation

Almighty news feed screen with categorised story cards

Almighty photo sharing screen with a multi image post

Almighty comment thread with one tap translation across languages

Already shipped

AI running in production, inside someone else’s product

Two client apps, both live, both built end to end by me, and my name is on neither listing. That is the arrangement you would be buying. The code is under NDA, the apps are not, so every claim below is checkable. Where you can confirm a feature from the public store page, it says so. Where you would have to install the app to see it, it says that instead. The translation is Google’s model through their API, and it is labelled that way rather than counted as mine.

AlmightyIn the stores

A social and messaging app with chat, voice and video calls, communities and a feed. React Native, iOS and Android, updated continuously.

App Store Google Play

AI I built for it

Inside the appTone check before a message sendsA message that is going to land harsh gets caught on the way out, and the sender is asked whether they want to soften it.
Inside the appTone rewriteThe softer version, generated, with the meaning kept. The sender picks which one goes.
Inside the appSensitive information detectionFlags things like a phone number or an address before they leave in a message the sender did not think twice about.
Inside the appCrisis language detectionSelf harm signals in a message are detected and surfaced. It is a safety net inside the product, not a clinical tool, and it is built and described as one.

Wired in, not built by me

On the store listingTranslation on every screen, 50+ languagesAny post, comment or message rendered in the reader's own language. The translation model is Google's, through their API. What is mine is the integration: wiring it into every surface in the app and making it hold up under real traffic. It is on the store listing and has its own screenshot.

These are the screenshots on the App Store and Google Play listings. Open one and check.

EzeeNotesIn the stores

A study app for Class 11 and 12 students. Notes, concept videos behind QR codes, and practice questions. React Native, Android.

Google Play

AI I built for it

Inside the appPersonal study analyticsEach student's own activity turned into a read on what they actually know and what they keep avoiding, generated per student rather than pulled off a fixed dashboard.

These are the screenshots on the Google Play listing. Open it and check.

What I build

Four things, not everything

Each of these ends as a feature your users touch, inside the product you already have.

An assistant over your own data

Your documents, your tickets, your catalogue. It answers from what you actually have, says so when it does not know, and lives inside your product rather than beside it.

Voice, vision and translation

Speech in and out, reading images and documents, translation that runs on every screen. This is the category already shipped and running in the stores.

Agents that do the work, not the demo

Multi step jobs with real tool access, running in the background against your systems, replacing something a person is currently doing by hand.

AI inside an app you already shipped

You have a live React Native or native app and a feature that keeps slipping. I go into your codebase and put it in, through your review process and out through your release.

Fit

When this goes well

The guarantee only works if we both know what we are aiming at. These are the conditions that make that possible.

You have a product with users in it, and a feature that keeps sliding to the next quarter.
You want the feature inside the product, not a demo living next to it.
Someone on your side can say, in one sentence, what working looks like.
You would rather look at a build than read a proposal.

No spin

What I am, plainly

Who you get

Mohammed Zaiban, working on it directly. LinkedIn. Not a company and not a team. Where this page says we, it means you and me.

Can you see the code

No. Both apps belong to clients and the code is under NDA, and so is anything I build for you. The same thing that stops me showing you their code is what stops the next one seeing yours.

How many at once

One build at a time. While one is running the next one waits, which is the honest reason I care what a project is before I take it.

Where I am

Outside the US. My hours overlap US afternoons, so you get answers inside your working day.

Whose name is on it

Yours. The two apps above are client work and my name is on neither store listing, which is the same arrangement you would get. I am not on the call and I am not in the deliverable.

What it costs

A project rate, once we both know what the thing is. Ask in the email and you get a number back.

Worst case, it costs you nothing.

Send the product, the feature you want in it, and what working would look like. Three lines is enough. If the build is not right you pay nothing, so the only thing you are risking by writing is the time it takes to write.

zaiban@greeencode.com

Prefer a form? Use the contact page. It lands in the same inbox.

zaiban@greeencode.com

You only pay if you want to keep it.

How you end up paying nothing

We agree what working means

I build it and you watch it run

You decide, and one option costs you nothing

AI running in production, inside someone else’s product

AlmightyIn the stores

AI I built for it

Wired in, not built by me

EzeeNotesIn the stores

AI I built for it

Four things, not everything

An assistant over your own data

Voice, vision and translation

Agents that do the work, not the demo

AI inside an app you already shipped

When this goes well

What I am, plainly

Worst case, it costs you nothing.