AI Tech InstituteAI Tech Institute
AI Tech Institute
Outcomes  ·  what you leave with

You don't leave with a certificate. You leave with something that runs.

Every program here ends in a real artefact, built on your own work and defended in front of the room. That artefact is the outcome. This page shows exactly what it is for each program, how it is earned, and the claims we refuse to make.

Programs
12
Every one ends in a defended artefact
Cohort size
12–25
Capped, so the defence is real
Assessment
The work
Not an exam, not attendance
Borrowed statistics
Zero
Method published or nothing said
001  — The artefact

One program, one thing you can show.

This is the full ledger. Every row is a real deliverable, written into the program before anyone enrols — not a learning objective, a thing that exists at the end and has your name on it.

Beginner AI Foundations
A project you built and can explain.
8 weeks
foundation
·
15
seats
AI & Data Science
An argument you can defend — including what it does not show.
12 weeks
intermediate
·
12
seats
AI Engineering & Production
A deployed, monitored service you built and can operate.
12 weeks
intermediate
·
12
seats
Data Engineering
A production platform you built, priced, and can defend end to end.
12 weeks
intermediate
·
12
seats
Generative AI Systems & Safety
A system that survives a red team — including the attacks that worked.
10 weeks
advanced
·
12
seats
AI Automation & Agent Builder
An agent running against your own tools, defended under questioning.
8 weeks
intermediate
·
12
seats
AI for Product Managers
A scoping case you defend — to build the feature, or to refuse it.
6 weeks
intermediate
·
15
seats
AI for Entrepreneurs
A ninety-day plan you'll actually run, with costs and a measure of success.
6 weeks
foundation
·
15
seats
Applied AI Capstone
Three production-grade projects, built and defended.
24 weeks
advanced
·
12
seats
AI Leadership & Governance
A ninety-day plan written in the room, for your organisation.
Half-day to 2 days
executive
·
15
seats
Building with Claude
One real change shipped from your own repository, that day.
1 day
foundation
·
20
seats
Copilot at Work
A working prompt set built around the tasks you actually repeat.
Half day
foundation
·
25
seats

Workshops produce an artefact you take away. Cohorts produce one you defend. The difference is the room: in a capped cohort there is time for every person to be questioned on their own work, and that questioning is where the learning lands.

002  — How it's earned

Built on your work. Defended in the room.

Week one

You bring the real thing

Your repository, your dataset, your process, your feature. We do not hand out a sample problem — the whole program is built around the thing you actually need to work on.

Through the middle

You build it under supervision

Live sessions with a senior practitioner who has shipped this before. Feedback on your work, not on a model answer. The system gets harder questions asked of it each week.

The last fortnight

You defend it

You present what you built to the cohort and your instructor, and answer for every decision inside it. Not a demo — a defence. That is the assessment.

The standard

We ask everyone for the post-mortem. The ones who have nothing to report usually haven't pushed hard enough.

003  — On Monday

What actually changes when you get back to work.

Capability, stated as things you can now do. Every line below is lifted from a program's published outcomes — not written for this page.

If you engineer

You can operate it, not just build it

  • A deployed, monitored AI service you built and can operate
  • Evaluation practice that tells you whether a change was an improvement
  • Guardrails, fallbacks and incident handling for systems that fail quietly
AI Engineering & Production →
If you analyse

You can defend the answer

  • Take a question and a dataset to a defensible, evaluated answer
  • Statistical judgement — knowing when a result is real and when it is noise
  • A working mental model of how language models and retrieval function
AI & Data Science →
If you decide

You can read the evidence yourself

  • Judgement about which AI features are worth building and which are not
  • The ability to read an evaluation and challenge it credibly
  • Working familiarity with the AI6 practices, in operational terms
Product & leadership programs →
If you run the business

You can hand over the right work

  • Automations running in your own business, not in a sandbox
  • A clear map of which tasks to hand over and which to keep
  • A way to judge the next tool pitched at you, without a demo
AI for Entrepreneurs →
004  — How we report

Method first, number second.

We are a young institute. Rather than borrow an impressive-sounding statistic, we publish the way we measure — so that when the numbers come, you can check them instead of trusting them.

i

Method published alongside

Any number we publish carries its definition, sample size and cohort year. No figure floats free of how it was measured.

ii

We measure what matters

Did you finish. Did you ship something real. Can you do the work now. Completion, capstone shipped, and capability assessed before and after — by you and by your instructor.

iii

Consent, always

Every named graduate and quote on this site appears with explicit permission. No composites, no invented voices.

005  — The line

What we will not claim.

The absence of these is itself a signal. Every one of them is standard practice somewhere in this market.

✕Guaranteed salaries, or a “job in 90 days” we cannot substantiate.
✕Placement rates without the methodology and sample size beside them.
✕Borrowed enterprise logos from work we have not genuinely done.
✕Accreditation we do not yet hold — we are not a Registered Training Organisation, and we say so.
✕Invented or composite testimonials. Real, named people, or nothing.
✕Cohort sizes or outcomes inflated to make us look bigger than we are.
006  — Graduate voices

In their words, with consent.

“The AI & Data Science program was one of the most impactful learning experiences I've had. Well structured, with a strong balance between theory and hands-on labs.”

Ehsan · Graduate

“Small cohort, real instructor, honest feedback. It felt nothing like the online courses I'd bounced off before.”

Pantea · Graduate

“They told me which course I actually needed — not the most expensive one. That's why I trusted the rest.”

Maria · Graduate

Quotes shown with permission. As cohorts complete, fuller outcome figures will be published here — with the methodology attached.

Judge us by the artefact.

Bring the thing you actually need to build. On a thirty-minute call we'll tell you which program produces it — or that none of them do, and what to do instead.