Training · AI-native employees · Operating systems

AI-native employees for every department. One operating system runs them. Your team, trained to lead them.

We start by training your team. Then we build the employees each department needs, run them with you, and hand everything over. They do the work no one can, or wants to, do. Your people keep the judgment.

Start with training Book 15 minutes

Built on Claude · Registered member of the Claude Partner NetworkWe run SMOrchestra on the same architecture: 100+ operating systems, 900+ skills, 39 live tool connections

Not sure where to start? Book the one-day audit.

An AI-native operating system at the centre, running six AI-native employees: a personal assistant, a social media manager, a sales nurture specialist, a data analyst, a video editor and a finance clerk, with finished work flowing back to the centre.

Brands we work with

Brands we work with.

From real-estate developers and construction to healthcare, higher education, B2B and training.

  • Uniphore logo
  • mdr logo
  • Princess Sumaya University for Technology logo
  • Taqniyat logo
  • CX-Dubai logo
  • mdr construction logo
  • Specialized Medical Center logo
  • Graia logo
  • Dtec, Dubai Technology Entrepreneur Campus logo
  • ICXI logo
  • mdr real estate marketing logo
  • Arcet Global logo
  • LAB experience logo
  • New Mind logo
  • mdr property management logo

Uniphore mdr Princess Sumaya University for Technology Taqniyat CX-Dubai mdr construction Specialized Medical Center Graia Dtec, Dubai Technology Entrepreneur Campus ICXI mdr real estate marketing Arcet Global LAB experience New Mind mdr property management

Anthropic research · March 2026

We take your business from the red to the blue.

Anthropic's radar chart of 22 occupational categories: a large blue area shows the share of tasks AI could theoretically perform, a much smaller red area shows observed AI coverage in real use
Blue is what AI could already do in each job family. Red is what people actually use it for.
90%

of office and admin tasks AI could take on, in theory

31%

covered in real use, on average, across office and admin, business and finance, sales, and computer and math

Same model for everyone. The difference is what you build.

Anthropic, “Labor market impacts of AI: A new measure and early evidence”, 5 March 2026, report p.7 and appendix p.6

What your company looks like today

Who holds it all together?

Today, you are the glue.

Tomorrow, the AI-native operating system is.

Same tools. Same people. One centre. You go back to decisions.

What an operating system is

Software is something you log into. An operating system is how the work gets done.

InputA lead, a request, a signal
The operating system
  1. SOPsHow each job is done
  2. StandardsWhat good looks like
  3. OrganisationOwners, cadences, decision rights
  4. Autonomous employeesRun on schedule
  5. Agentic employeesRun when triggered
  6. Your peopleDecide, approve, handle exceptions
OutputFinished work, scored, one named owner
Agentic

You trigger it. The employee does the work; you approve before anything leaves. For judgment, money or reputation.

Autonomous

It triggers itself. Runs on a schedule, scores its own output, escalates exceptions. For repeatable work with clear rules.

The OS comes first. Employees are who you hire into it.

Runs on Claude, wired into the CRM, WhatsApp and tools you already use.

Your AI squad · inside the operating system

One operating system. A squad of AI-native employees.

Your AI-native OSOne brain · four employees

  • Personal Assistant

    Buys back your time

    • Inbox triaged, replies drafted
    • The morning brief
    • The weekly report

  • Social Media Manager

    Gets you seen

    • Posts in your voice
    • Carousels and infographics
    • The weekly posting plan

  • Video Editor

    One recording, many clips

    • Clips from one recording
    • Captions and thumbnails
    • Hooks scored before posting

  • Data Analyst

    Turns files into decisions

    • Data cleaned and checked
    • Your questions, in numbers
    • A dashboard and a decision brief

Same model under all four. Different brain, skills and standards.

In your industry

  • Real-estate developers and construction

    Social Media Manager · Enquiry follow-up on WhatsApp · Project reporting analyst

  • Healthcare

    Patient booking coordinator · A clinic operating system

  • B2B

    Outbound specialist · Account research analyst · Proposal writer

    See the B2B revenue engine

  • Training companies

    Enrolment follow-up · Content engine · Webinar operations

The five levels · From fresh graduate to a department in a cockpit

The five levels of an AI-native employee. The command gets shorter as the employee gets built.

Same Claude at every level. The level is not a trait of the model; it is how many layers you built underneath it before you typed.

L1Executes

A fresh graduate. Works on your files, but you explain everything, every time.

The command you give
“Read this folder, write a summary, this format, this length, focus on that sector, avoid this, that tone, then a table, and the table should…”
Three paragraphs, every time
A draft you correct.
The free session (9 Dec)
L2Repeats

Learned your way of doing one job, and now repeats it right without being reminded.

The command you give
“Prepare the weekly report, same format we agreed, with this week's numbers.”
Two lines
The same shape every time, unprompted.
Master Claude, $297
L3Works your tools

Skills installed, your tools connected. It opens, reads and prepares inside them.

The command you give
“Prep tomorrow's meeting.”
Three words
Finished work inside your tools. A file, not advice.
Build Your AI-Native Employees, $1,750
L4Your employee

One job run end to end: your skills, your folder, your tool keys, in concert.

The command you give
“Run the insurance campaign end to end.”
One sentence
A campaign running inside your tools.
Build Your AI-Native Revenue Engine, $6,500
L5A department

Many employees, each with its skills, one orchestrator running them from one cockpit.

The command you give
“Open Saudi.”
Two words
A department running. You decide.
Built for you: the revenue engine or the dev department

Same model in all five. The difference is how many layers you built under it.

The three macro operating systems

Three operating systems run a company.

Your companyOne brain · three operating systems

Same pattern inside every one. Input, the system, scored output.

Proof · what we have built

Not slides. Employees and operating systems, running.

Every one below was built in Claude, wired into real tools and run with a real team. The screens are captures of them working.

18

AI-native employees

Marketing and content

  • Social Media ManagerBuilt for: Real estateHealthcareFounders
  • SEO and GEO SpecialistBuilt for: Real estateB2BFounders
  • Marketing Manager
  • Campaign ManagerBuilt for: Training
  • Video Editor
  • Product Video GeneratorBuilt for: B2B

Revenue

  • Outbound Revenue EngineBuilt for: B2B
  • AI-Native Outbound EngineBuilt for: B2B
  • Nurture GTM SpecialistBuilt for: CoachingB2BE-commerce
  • AI-Native Ads EngineBuilt for: CoachingTrainingReal estate
  • Sales OpsBuilt for: B2B

Operations and service

  • Booking AssistantBuilt for: Healthcare
  • Support AgentsBuilt for: Healthcare
  • HR Manager
  • Data Analyst

Leadership

  • Personal AssistantBuilt for: Leaders
  • Business ManagerBuilt for: Leaders
  • Co-FounderBuilt for: SMEs
7

Operating systems

Revenue

  • Signal Inbound Engine (SINE)Built for: Real estateB2BFounders
  • Signal Outbound Engine (SOE)Built for: B2B
  • Signal Ads Engine (SAE)
  • Sales Operations EngineBuilt for: B2B

Engineering

  • AI-Native Engineering for MicroSaaS
  • AI-Native Engineering for AI Agents

Business

  • SME Operating SystemBuilt for: SMEs

Screens, captured running

The CEO Brief email: a morning summary that names every system it read, then open opportunities, warm replies and channel results, with the figures blurred
The CEO Brief: what the Personal Assistant hands its leader every morning (figures blurred)
The Social Media Specialist plugin detail page: a publish-ready week in brand and voice, creatives sized per channel, copy past an anti-AI writing gate
The Social Media Specialist: a publish-ready week in brand and voice
The Sales Nurture Specialist plugin detail page: every lead bucketed cold, warm or hot, follow-up drafted, nothing sent until the founder confirms
The Sales Nurture Specialist: every lead bucketed, follow-up drafted, nothing sent without a yes
The SINE product: declared inputs with one identifier redacted, ten channels on a weekly publishing calendar, and an autonomy dial locked to approve-first that loosens per channel after ten clean approvals
SINE, the Signal Inbound Engine: ten channels on one weekly plan
The SOE product: a review queue, signals scored and tiered before anything sends, and campaigns and sequences behind it
SOE, the Signal Outbound Engine: signals scored and tiered before anything sends
The SalesMfast product on a demo workspace: opportunity status, value by stage, conversion rate and the funnel by pipeline
Sales operations on SalesMfast: pipeline value, stages and conversion
The AI-native dev org as a company chart: a chief executive, a chief marketing officer with a content lead, a GTM specialist and product marketing, a VP of engineering with six engineers, and a QA lead, every seat a running agent
The AI-native engineering org: a company chart of agents
Where every engagement starts

Every engagement starts with training.

Your people learn to direct AI-native employees before we build them. Every program ends with a working system, not a certificate.

Start hereTraining first, then the build
The last use case

“Let’s hire AI to replace our staff.”

That’s the last use case. Start with these.

  1. 01

    What are you procrastinating?

  2. 02

    Freed from it, what would your team focus on?

  3. 03

    What do you wish was off your to-do list?

  4. 04

    Which revenue work never gets enough hours?

  5. 05

    Which chores does everyone hate?

  6. 06

    Where do you hire agencies?

  7. 07

    What are you outsourcing today?

  8. 08

    Where do things get lost: communication, comprehension, translation?

  9. 09

    Where is orchestration hardest?

Your first AI employee is hiding in these answers.

Book 15 minutes
Build · Operate · Transfer

They rent you capability. We transfer it.

Every engagement has an end date. We build, we run it with your team, then we hand over the whole system: accounts, code, configs, data. Transfer happens inside the last weeks of operation, while it is still running, with us in the room.

vs. Agency
Pay forever. Own nothing.
The day you stop paying, the work stops too. We flip it: we build, we run it with you, then we transfer the whole system, accounts, code, configs, data. You keep a light retainer if you want one. Not because you need it.
vs. SaaS
The tool isn't the job.
Most CRMs end up as expensive address books. Not because the software is bad, because nobody operates it. We build the operator inside the tool, already trained on your business, so the CRM becomes the smallest part of the value.
vs. Consulting
A deck doesn't send an email.
You've been handed the 47-slide strategy deck before. It's still in your drive. Our deliverable sends emails at 3am, moves your pipeline stages, scores your copy, and lands reports in your inbox. Strategy that executes itself.
vs. Hiring
Three months to ramp. Eighteen to leave.
The salary is only part of the bill. The engine installs faster, costs less, and does not walk out with what it knows.

Built on Claude. The model is becoming a commodity. The operating system is what you own.

Ways in

Three ways in. Every price published.

No founding pricing, no introductory rate, no "contact us for a quote". If a number is on this page it is the number on the contract.

01 · TrainingYour people learn to direct the squad before we build it.Individuals from $297 · companies from $5,000Start with training 02 · AuditRevenue and GTM Audit or AI Automation Audit.$1,500 · one day · credited in full against the matching build within sixty daysSee the audits 03 · BuildAI-native employees and operating systems, built, operated, transferred.Revenue OS $7,500 per process · Expert Business Engine $16,000 · B2B Revenue Engine $39,000See every price

What we need from you: four hours a week and one decision-maker who can say yes without a committee.

The Operator Behind the Engine

I've sat in every seat this engine replaces.
That's the whole reason it exists.

Twenty-two years inside Google, Cisco, Avaya and Uniphore, then VP Sales and founder-operator work for my own bets: the same stack MENA founders are being sold as "transformation", enterprise sales, RevOps, GTM for B2B SaaS, expansion into Saudi and the UAE. I was the VP of Sales watching CRMs go unfilled. The head of GTM hiring SDRs we couldn't actually retain in Dubai. The founder quoting a client a six-figure rebuild we both knew would slip by four months.

At some point it stopped being a people problem. The roles existed. The playbooks existed. The talent, on the Western side, was plentiful. What didn't exist was a way to install that capability into a MENA-based founder's business without first winning the North American hiring war.

An AI-native revenue engine is the answer I wish I'd been able to buy fifteen years ago. It's the way we finally stop telling MENA founders they need to hire a Riyadh SDR who doesn't exist, or retain a Dubai RevOps lead who'll leave for Careem in six months. The operating system is the hire. The transfer makes it yours.

22
Years in GTM, RevOps, and founder-operator roles across MENA and global SaaS.
KSA · UAE
Expansion work across the two Gulf markets this OS is engineered for.
4 engines
Two live, one transferring, one in build. The status of each is on its card below.
LinkedIn · Mamoun [email protected] Book a call with me
The Honest Part · Fit Check

This isn't for everyone. Here's how to tell.

We'd rather you self-select out in 30 seconds than ten weeks in. The OS only compounds if the founder is ready to operate it, not just buy it. Read both columns. If one feels more true than the other, you already have your answer.

This works if you're…
  • A founder or operator doing five jobs at once and losing on all of them.
  • Past founder-selling and above $500,000 in annual revenue, with a sales team of two to twenty.
  • Operating in MENA or selling into it: real-estate developers and construction, healthcare, B2B and training companies. WhatsApp, dialect and trust-first buying are non-negotiable.
  • Willing to be in the room for the twelve weeks while we build the engine with you, not just for you.
  • Serious about owning the stack, accounts, code, configs, end-state, not rental.
Skip this if you're…
  • Looking for a cheaper agency retainer, we charge for the transfer, not the dependency.
  • Below the $500,000 revenue floor. Start with the $6,500 cohort instead, a motion has to convert before it is worth building an engine around it.
  • Expecting a SaaS with a login, a pricing page, and a 14-day trial, there's no product to subscribe to.
  • Unwilling to touch the tools, you want AI that runs while you never learn what it does.
  • Looking for someone to hand a deck to and disappear, this is build-operate-transfer, not a handoff.
What this is not

Not a RevOps consultant. Not a SaaS. Not a retainer.

Every category someone tries to slot us into gets the shape wrong. A revenue engine you own sits somewhere new, and that's the whole point.

Not a RevOps consultant

Consultants ship slide decks and leave you with an even bigger to-do list. We ship installed systems you own and operate after we walk away.

Not a SaaS tool

There's no login, no seat pricing, no feature flag gate. The tools are yours, Claude, n8n, Supabase, Stripe, all in your name, your accounts, your control.

Not an agency retainer

Agencies bill for hours you don't see. We bill for a build, operate it beside you, then transfer it. The optional retainer after that is maintenance, not survival.

Here's what we actually are

A build-operate-transfer team that installs one of the three pillars on Claude-native infrastructure, and then hands you the keys. Once. Properly.

Questions

Six questions, answered straight.

What is an AI-native employee?

An AI that holds a job, not a chat: a role, a brief, tools and a scorecard. It does the work, and someone on your team checks it before anything leaves.

What is an AI-native operating system?

The centre every AI-native employee runs on: memory, skills, quality gates and one scorecard, wired into the CRM, WhatsApp and tools you already use. One operating system runs the employees of every department.

Where do we start?

With training. Your team learns to direct AI-native employees before we build them, and every program ends with a working system, not a certificate. Then we build the employees each department needs and the operating system that runs them.

Do we need Claude?

Yes. Claude is our core: every employee and operating system we build runs on it. We wire it into the tools you already run: your CRM, WhatsApp, email and files.

Who owns it at the end?

You do. We build it, run it with your team, then transfer the whole system: accounts, code, configs and data. Every engagement has an end date and no retainer.

Which industries do you build for?

Real-estate developers and construction, healthcare, B2B and training companies, plus coaching, e-commerce, founders and SMEs. The proof section above shows what we built for each.

Where this goes

The future of growth is ANB.

ANB. The AI-native business.

AI does the work no one can, or wants to, do. People do the work only people can do: judgment, relationships and decisions.

Here is our AI organization.

The real org chart of SMOrchestra's AI-native development company: thirteen AI agents, a chief executive named Sulaiman, three leads and nine specialists, each running as an agent12345678910111213
13

roles on one board.
Every one is an AI agent.

1

human directs it, from Telegram.

This is our development company, running today. Every cockpit reports to Sulaiman, the AI CEO, and every role is an AI-native employee on one operating system.

  • 01

    AI-native employee

    An AI that holds a job, not a chat: a role, a brief, tools and a scorecard. It does the work, and someone checks it.

  • 02

    AI-native operating system

    The centre every employee runs on: memory, skills, quality gates and one scorecard.

  • 03

    AI-native business

    A company built so AI employees run the routine work in every department, all reporting into one system.

  • 04

    The director: you

    You set the standard, approve what matters and make the decisions.

Book 15 minutes Free readiness scorecard, 5 minutes

Not ready to talk? Read how we build on the blog, or take one argument a week in the Tuesday letter.