Skip to content

Nine AI solutions. The diagnosis picks yours.

These are the nine AI solutions we implement inside a company. Every module on the list already has a version built and running somewhere, and that is the only reason it is here. No company buys all nine: the diagnosis points to the two to four that remove the most expensive bottlenecks, and those are the ones that get built.

What a build usually looks like

Nobody buys the menu. A build is the foundation, plus the two to four modules the diagnosis picked, plus the integrations those modules need to actually act, plus the handoff. That last part is not optional.

  1. 01

    The foundation first

    The company second brain goes in before anything else. Agents are only as good as what they know about your business, and everything above reads from it.

  2. 02

    Two to four modules

    The ones the diagnosis showed are costing the most. Not the ones that demo best, and never all nine.

  3. 03

    The integrations they need

    A module that cannot reach your CRM, your inbox or your documents is a demo. The connections are part of the build, not an extra.

  4. 04

    Handoff and training

    Your team trained, the system documented, and 30 days of tuning while they run it. Without this the company owns a system nobody can operate.

Foundation

The layer everything else plugs into

Agents are only as good as what they know about your business. The foundation comes first, then the things that run on top of it.

01

Company second brain

Your operation's knowledge in one place, structured for both people and AI: context, processes, decisions, history and documents, with permissions so each team sees what it should. The agents read from the same source your team does, which is what stops them from inventing answers.

Replaces

Knowledge scattered across drives, chat threads and the heads of whoever has been there longest.

Integrates with

The documents, drives and tools your context already lives in, with permissions per team.

Done means

Your people and your agents answer from the same source, and adding to it is part of the normal week.

Built on the AI OS core that runs my own operation every day

Agents

The work itself, done by agents

These are the modules that take repetitive work off expensive people. Each one runs with the approval gates you define, never unsupervised on anything risky.

02

WhatsApp and messaging agents

Sales, support or concierge on the channel your customers already use. The agent qualifies, answers, routes and books, and hands over to a person at the point you decide. Ports to Instagram, Slack, Telegram and web chat with the same core.

Replaces

A person retyping the same twenty answers, and enquiries that sit unanswered until someone opens the inbox.

Integrates with

WhatsApp first, and the same core ports to Instagram, Slack, Telegram and web chat.

Done means

The agent qualifies, answers, routes and books on its own, and hands to a person exactly where you drew the line.

Built on the agent that answers my own business WhatsApp in production

03

AI admin assistant

Email, calendar and documents with triage, drafting and scheduling done by AI. It proposes, you approve, then it sends. The gate is the point: nothing leaves the company without a human saying yes until you decide otherwise.

Replaces

Manual triage, drafting from scratch, and the calendar tetris that eats a morning.

Integrates with

Google Workspace: email, calendar and documents, through OAuth with an approval gate.

Done means

It proposes, a person approves, then it sends. Nothing leaves the company without that step until you remove it.

Built on a Google Workspace assistant with OAuth and an approval gate

04

Meeting intelligence

Calls and meetings become structured records, decisions, action items and scheduled follow-ups. The pipeline runs from recording to task without anyone writing notes, so commitments stop depending on memory.

Replaces

Someone taking notes instead of participating, and follow-ups that depend on remembering.

Integrates with

Your recorder, your task system and your calendar.

Done means

A recording becomes a structured note, decisions, action items and a scheduled follow-up, with nobody writing anything down.

Built on a transcript to note to scheduled follow-up pipeline in daily use

Automation

Routines that run with nobody driving them

The unglamorous layer that quietly removes hours: things that should happen on a schedule, happening on a schedule.

05

Internal automations

Background workers that route incoming information, summarise what changed, alert the right person, assemble recurring reports and keep data tidy. They run on a server on a schedule, report when something breaks, and need no one to remember them.

Replaces

Recurring manual routines: the weekly report assembled by hand, the alert somebody forgets to send, the data nobody tidies.

Integrates with

Wherever the information already lives, on a schedule, on your own server.

Done means

It runs unattended, it reports when it breaks, and no one has to remember it exists.

Built on unattended workers running on a cheap VPS, monitored and logged

Growth

Demand, built as a system

This is where the group's marketing side and the AI side meet. Same discipline as the internal work: agents do the volume, a person keeps the judgement.

06

Content pipeline

Research, drafting and review across channels, with a human editor in control of what ships. Volume without losing the voice, because the brand rules live in the second brain rather than in someone's head.

Replaces

One person being the bottleneck for everything that gets published.

Integrates with

The brand rules in your second brain, and the channels you already publish to.

Done means

Volume ships on schedule with a human editor deciding what goes out, and the voice still sounds like you.

Built on multi-agent content swarms and a library of roughly 65 skills

07

SEO and answer engine command centre

On-page, technical, Search Console and visibility inside AI answers, operated by supervised agents. Optimising for what ChatGPT, Perplexity and AI Overviews cite is now a separate job from ranking on Google, and this module treats it as one.

Replaces

Manual audits, and guessing whether you appear in AI answers at all.

Integrates with

Search Console, your site, and the assistants your buyers ask before they ask Google.

Done means

On-page, technical and AI-answer visibility are all operated by supervised agents, and you can see what changed and why.

Built on nine SEO and AEO skills used on live client sites

Software

When the answer is a product, not an agent

Sometimes the bottleneck is not a process that needs automating. It is a tool that does not exist yet.

09

Custom app or SaaS build

A customer portal, a booking system, an internal operations console, or a product you intend to sell. Designed, built, deployed and handed over, with payments, authentication and notifications where the case needs them.

Replaces

The tool that does not exist, and the spreadsheet standing in for it.

Integrates with

Payments, authentication and notifications where the case needs them.

Done means

It is designed, built, deployed and handed over, running with your users on it.

Built on a multi-tenant booking SaaS live with paying companies

One build, end to end

What this looks like when it is us

The nine above are a menu, and a menu is easy to write. So here is one of them built and running, on the operation we know best: our own. Same four steps a client gets, in the same order.

01

What it looked like before

Enquiries arriving on WhatsApp at every hour, the same twenty answers retyped by hand, and contacts sitting unanswered until somebody opened the inbox. Not a technology problem. A person doing work that did not need a person.

02

What the diagnosis found

The bottleneck was not volume, it was response time and repetition. Most enquiries needed one of a small set of answers plus a qualifying question, and the ones that genuinely needed a person were arriving mixed in with the ones that did not.

03

Which modules were picked

One: messaging agents. Not the second brain first, because the context already existed. Not the admin assistant, not the automations, not the content pipeline. The diagnosis pointed at one bottleneck and one module removed it, which is what two to four modules looks like when the answer is one.

04

What it does now

It qualifies, answers, routes and books on its own, and hands to a person exactly where the line was drawn. It runs in production every day on its own server. Done was never the launch: done was the day nobody needed to open the inbox to find out whether something had gone unanswered.

It is the module we sell most often, and it is the one that can be demonstrated live instead of described. Ask for that on the diagnosis call.

Three ways in, by the size of the company

The same work is shaped differently depending on who is in the building. An owner with no team does not need a champions cohort, and a company with departments does not need a session with the founder every fortnight. Pick the row you are on.

01

CompanyAn owner, with a small team or none

Eight sessions, one at a time

You are the operation. The sessions run with you, in plain language, no terminal and no code. Session two builds the document that holds everything about the company, and from session three each one kills a task that eats your week.

ShapeEight sessions, about an hour each

02

CompanyA company with areas and teams

The programme, in three layers

A shared standard for everyone, then each area working on its own repeated tasks, then a small internal team through the twelve weeks that keeps it alive. This is where day to day productivity across the company actually comes from.

ShapeThree layers, twelve weeks on the third

03

CompanyA company that wants the system built

Diagnosis, build, handoff

We find what the bottlenecks cost, build the modules that remove them, and hand the system over trained and documented. Training is inside it, at the handoff, rather than being the product.

ShapeAbout three months, four phases

You are reading this one

They are not alternatives to argue between. Most companies start on one row and end up on another, and the conversation that decides which is always the same one.

Questions about the solutions

Do the nine solutions already exist, or are they built from scratch?

Every module on the list already has a version built and running somewhere, and that is the only reason it is here. The diagnosis points to the two to four that remove the most expensive bottlenecks, and those are the ones that get built.

Do you use these solutions in your own operation?

Yes. The group's own operation runs on these modules: a second brain, around 65 skills, and agents working on a server all the time. The page walks through one of them built and running in the operation we know best, ours, in the same four steps a client gets.

What if I need an integration that was not in the proposal?

Integrations the diagnosis did not put in the build are a new quote, not a favour. The same applies to software beyond the modules the proposal named.

Start with an AI diagnosis.

Thirty to forty five minutes. We find out whether there is a bottleneck worth removing, and I tell you honestly if I cannot help.

Book a diagnosis