16 hours · 4 sessions
L1 — Get Your Company AI-Ready
Your staff learn to use AI safely on real company work, and the company ends Level 1 with a written AI use policy and its first working workflows.
The courses
All three levels and all 24 sessions, on one page
Three levels, 72 hours, 24 online sessions, four modules in every level. Level 1 for staff who need AI used safely on real company work, Level 2 to build the company's own agent — the software theirs, the machine ours — and Level 3 to build a customised project on a server the company controls and owns. This page is the whole syllabus, written from the sessions that are actually delivered.
Each session is worked online at the person's own pace, with a scheduled live touchpoint inside it, and each one ends with something your company keeps. The level pages carry the detail; the sessions themselves are delivered inside the course, to the people you enrol. Level 1 is running now. Level 2 is built: every session is written and running in the course platform, and we are not enrolling companies onto it yet - ask us and we will tell you the day it opens. Level 3 is not open yet: its sessions are still being written, and we will not sell a level we cannot hand your team to work through. Nothing is charged for a level until it opens.
Ask us for a free quotation for the seats you need — nothing is charged before you approve it.
72 hoursacross three levels
24 sessionsown pace, online
Onlinea live touchpoint in every session
On your worknot textbook examples
Three levels, one ladder
Each level, and what the company keeps
Level 1 is where a company starts: AI used safely on real work. Level 2 builds the company's own agent. Level 3 is a customised project the company builds and then runs on its own server. Most companies take one level, see it land, then decide about the next.
24 hours · 8 sessions · not open yet
L2 — Build Your Company's Own AI System
Your staff research the use case, agree what data may be used, then build and test the company's own AI colleague on one piece of real work.
32 hours · 12 sessions · not open yet
L3 — A Customised Project — Build, Monitor, Grow
One project chosen on the company's own numbers, then built deeper than Level 2: the company's data and documents on a server it controls — existing, rented, or bought for the job — with the paperwork automated beside it.
No hours sat through, no model answer
What each level is measured on
Progress is judged on the work someone produces, not on how long they sat in front of it. Every level ends with a live assessment on the learner's own work: On the work your team produces and can run without us: the workflow, the policy, the agent or the customised project that this level is built to produce — checked against your own real work, not a model answer.
L1 · session 4 · the assessment
L1 — the assessment
Your signed one-page AI-use policy, plus each person's live demonstration of their own working app on their own work
L2 · session 8 · the assessment
L2 — the assessment
A live demonstration on fresh inputs, and your company agent record
L3 · session 12 · the assessment
L3 — the assessment
A live demonstration of the customised project, and the record that goes with it
L1 · 16 hours · 4 sessions
L1 — Get Your Company AI-Ready: all 4 sessions
| # | The session | The one thing it settles | What your company keeps |
|---|---|---|---|
| 1 | What AI is, what it may be trusted with, and your first working app | What this tool can and cannot be trusted with, how to get usable work out of it on the first or second try, and how to describe a small app so the AI builds it | A one-page note in the person's own words on what AI is for at their desk, plus a small working app of their own — one screen, the work behind the button, and a store that keeps the work |
| 2 | Finding the work worth doing | Which three tasks in your company should come first, and which should not be touched yet | A shortlist of three tasks with the minutes per week each one costs and the risk noted |
| 3 | Data and documents, and catching a confident wrong answer | What may go in, what must never, and how a plausible but wrong answer is caught before it reaches a filing | Your draft do-not-enter list, written by the people who handle the documents, plus the written three-check habit and the log of what it caught on real work |
| 4 | The policy, and the assessment | What the company has decided, who is allowed to do what, and whether each person can actually do it | Your signed one-page AI-use policy, plus each person's live demonstration of their own working app on their own work |
The full L1 outline, with the assessment and the record it produces: level-1.html
L2 · 24 hours · 8 sessions
L2 — Build Your Company's Own AI System: all 8 sessions
| # | The session | The one thing it settles | What your company keeps |
|---|---|---|---|
| 1 | Choosing the right job for an agent | Which workflow is worth automating first, and which must not be | A scored shortlist of three workflows, and the one chosen, with the reason |
| 2 | Mapping the workflow and the data | Where the information lives, who owns it, and what may not be touched | A one-page workflow map plus the data rules for this agent |
| 3 | Writing the agent's brief | What it is told to do, and the three limits it may never cross | The agent's written brief, including the three limits |
| 4 | Building it | A first working version connected to your own information | A first build that answers correctly on three real inputs from your work |
| 5 | Testing it against real work | What breaks, and how you know | A failure log built from your own documents, with the fix for each failure |
| 6 | Guarding it | Who may ask it what, what needs a human, and what gets logged | The guardrail document, plus the corrected rule from the incident drill |
| 7 | Putting it into the daily routine | Who operates it, and what working means on a Monday morning | A one-page operating routine with a named owner and a reviewer |
| 8 | The build assessment | Whether your company actually has a running agent | A live demonstration on fresh inputs, and your company agent record |
The full L2 outline, with the assessment and the record it produces: level-2.html
L3 · 32 hours · 12 sessions
L3 — A Customised Project — Build, Monitor, Grow: all 12 sessions
| # | The session | The one thing it settles | What your company keeps |
|---|---|---|---|
| 1 | The project, chosen on the company's own numbers | Which work to fix first — the administration that eats the most hours, or a new line of income, with the system bringing back about three feasible ways to add it | A one-page project case with its numbers, the person who owns it, and the researched options beside their figures |
| 2 | The boundary, and where the data will live | What the system never decides, who each exception goes to, and the hosting choice: an existing server, a rented one, or one bought for the job | The boundary page signed by the process owner, and the written hosting decision with its reason |
| 3 | The machine, specified and quoted | What the hardware must be for this job, where it sits, its power and backup, and who may touch it | The server specification, the quotation request, and the equipment register |
| 4 | Install day: the model running on your own machine | The company's AI running on hardware it controls, with nothing leaving the building | The running system, its model card, and the written access list |
| 5 | The doors, the locks, and the team's way in | Who may reach it, over what, and the lock-down in order: two-factor, keys not passwords, three ports, the model never exposed to the internet, updates, sign-ups off, and a private network for staff | The locked-down configuration sheet, the access list, and a backup restored once to prove it |
| 6 | The company's own documents, and answers that cite the paragraph | Which papers go in, scanned files included, what is never loaded, and how an answer shows the exact text it came from | The document collection with its do-not-enter list, and the citation test sheet with its pass line |
| 7 | The paperwork that files itself | Which recurring jobs the system takes first, and what still needs a person | The automation list, each with its trigger, its owner, and the first date it ran |
| 8 | The four jobs we suggest first | The jobs a private AI does for a company at once: the weekly report written for the team, the contract form that pulls out the key dates, the support inbox sorted with a drafted reply, and a question answered from the company's own papers | The first two running end to end, each with a real sample of it working, and the rest listed with owners |
| 9 | Guarding a system that is entirely yours | With no outside provider holding any of it, the company carries every rule: the approval gate, the never-list, and the person told when it fails | The in-house guardrail document, the incident drill report, and the one rule that changed because of it |
| 10 | The daily routine and the watch log | The ten-minute check, who reads the log, and what a bad week looks like before it becomes a bad month | The watch log with its first entries filled in, and the reader's name |
| 11 | Records that stand up, and what it gave back | What a client, an insurer or an auditor would need, and whether the hours promised in session 1 came back | The evidence pack, answerable by a single named date, and the before-and-after on the company's own numbers |
| 12 | The customised project, demonstrated live | Whether the company can run, defend and maintain the system it built | The project record, and the assessment result part by part |
The full L3 outline, with the assessment and the record it produces: level-3.html
A free quotation
Ask us for a free quotation
Ask us for a free quotation for the seats you need; the seats and the starting level decide what it covers. What is free is the first step — the AI Readiness Report. The number of seats and the starting level are confirmed at the build assessment, because that is what decides the work.
Ask us for a free quotation for the seats you need · Start free with the AI Readiness Report
Questions about the path
The short answers
Can we start at Level 2 or Level 3?
Not yet. Level 1 is running now. Level 2 is built: every session is written and running in the course platform, and we are not enrolling companies onto it yet. Level 3 is not open yet: its sessions are still being written. We will not take a booking for a level we cannot hand your team to work through, so nothing is charged for a level until it opens. The build assessment is free and still decides the right starting point; if it places your team on Level 2 or Level 3, we will tell you the moment it opens, before you are asked to pay anything.
Do we have to buy all three levels?
No. Most companies take Level 1 and decide about the next level only after they have seen the first one land. One level that works beats three that sit unused, and that is the fair way round to sell it.
Is the lesson material on this page?
No, and deliberately so. This page is the syllabus: every session, what it settles and what it leaves behind. The sessions themselves are delivered inside the course to the people you enrol, and every learner's own record and progress are theirs.
What is kept per person, and who may read it: the record page.
What does the company keep at the end of each level?
L1: a signed one-page AI use policy and the first working workflows. L2: the agent itself, its written brief with the three limits, and a guardrail document. L3: the project record — the system specification, its records and the evidence pack. The record of what each person did is kept per learner — identity, attendance, progress with the work produced, and the assessment result part by part.
How is a learner judged?
On the work they produce, not on the hours they sat through. Every level ends with a live demonstration on the learner's own work: at L1 a workflow they run themselves, at L2 a demonstration on fresh inputs, at L3 a live demonstration of the customised project. The fair test of a training course is what someone can do the week after it ends.
Next step
Get the free AI report
A short online conversation: we look at the work eating your team's week, say plainly whether this fits, and quote for the seats you actually need. If it is not a fit, we will tell you on the call — that is the fair answer to someone deciding about their team's week.
Ask us for a free quotation for the seats you need — nothing is charged before you approve it.