— AI certification for data scientists —

Prove the judgement,
not just the models.

A good model is the easy part to show. What a portfolio cannot prove is whether you framed the problem well, handled the data honestly and can answer for the result. The Institute of AI is the UK’s professional body for AI, and IoAI accreditation assesses your practice across the whole data-science workflow, independently and against a published framework.

Find your level, free See the framework

Independent and vendor-neutral
Four levels, one framework
£29.99 assessment

— The gap a portfolio leaves —

Anyone can show a good model.
Few can prove the practice.

Hiring managers have seen enough polished notebooks. What they cannot read from a repository is the judgement around it: the problems you chose not to solve with AI, the leakage you caught, the metric you argued against. That is what accreditation makes legible.

What a portfolio shows

The results you were happy with, presented the way you wanted. Useful, but self-selected, and silent on everything that went into the choices behind them.

What accreditation proves

That you practise to a standard across the whole workflow: framing, data, modelling, evaluation and accountability. Independent, portable and verifiable, it travels with you between employers.

— Mapped onto how you work —

The framework, laid over your workflow.

The competency framework is not an abstract syllabus. Read across the stages of a real data-science project and you can see exactly where you sit and what the assessment probes at each one.

Read the framework

Framing and judgement

Where a problem should meet a model

The work a portfolio rarely shows: deciding whether a question should be answered with AI at all, scoping it honestly against the data you actually have, and knowing when a model’s confidence outruns its competence. This is where good data science is won or lost.

Translating a business question into a tractable modelling problem
Choosing metrics that reflect the decision, not the demo
Reading the limits of the data before promising an outcome

The assessment probes how you frame a problem, set success measures, and defend the call to build or not to build.

Data

The foundation everything rests on

Sourcing, cleaning, leakage, drift and the quiet assumptions that decide whether a result is real. Assessed as practice, not trivia.

How you evidence data quality, provenance and the checks that keep a pipeline honest.

Modelling

Building the thing well

Selecting, training and tuning models with a reason for each choice, and the discipline to prefer the simplest approach that works.

How you justify approach, baseline honestly and avoid overfitting the story to the sample.

Evaluation and deployment

Proving it works, then keeping it working

Measuring what a system actually does rather than what the notebook suggests, then getting it into production and watching it hold up. Evaluation, monitoring and regression are graded as first-class skills, because a model that degrades unnoticed is worse than no model at all.

How you evaluate under realistic conditions, ship responsibly and catch a model slipping before your users do.

Responsibility

Answering for the decision

Governance, data protection and accountability under UK expectations, including the ICO. Who owns the outcome, what leaves the machine, and how you defend it later.

How you handle fairness, privacy and the paper trail a regulator or a board would ask for.

— One credential, four levels —

One standard your career can grow into.

You apply at the level your current practice supports, and progress upward as it deepens. Professional is the grade for data scientists who own their work end to end and answer for the outcomes.

Student

Studying, or moving into data science. Establish the fundamentals before habits harden.

Associate

Applying data science in the day job, accountably and with growing independence.

03Independent practice

Professional

For the data scientist who owns their work end to end. Applied judgement and delivery, evidenced.

Fellow

Senior practitioners shaping how their organisation and the field do data science.

Find your level, free See the framework

— Why it holds up —

A standard set by
the profession.

A credential is only worth what stands behind it. The Institute of AI is the UK's professional body for AI, and its remit is the standard rather than any one supplier's stack. The competency framework is published before you apply, the assessment is judged against it, and the result is verifiable by anyone.

Apply for accreditation

A standard set by the UK’s professional body for AI, independently of any tool vendor.
A competency framework published in full before you apply.
Four levels from Student to Fellow, assessed against one standard.
Annual CPD targets by level, logged and evidenced.
Public verification for every accredited grade.

— Common questions —

Before you apply.

How is this different from a data science certificate or my portfolio?+

A course certificate proves you finished a syllabus, and a portfolio shows what you chose to show. Accreditation from the Institute of AI is independent and evidence-based: it assesses your practice across the whole workflow against a published competency framework, so an employer is trusting IoAI’s standard rather than taking your word for it.

I am a data scientist, not an engineer. Which level fits?+

Professional is the grade for data scientists who own their work end to end and answer for the outcomes. You apply at the level your current practice supports, and progress upward from there. The framework is discipline-agnostic: it recognises applied competence and judgement, whether you spend your days in notebooks, pipelines or production.

What does the assessment actually involve?+

An evidence-based assessment of your real practice against the competency framework, not a multiple-choice question bank. It looks at how you frame problems, work with data, model, evaluate, deploy and answer for the outcome, at the level you are applying for. Applications are reviewed against the framework alongside references you nominate.

What does it cost?+

A one-off assessment fee of £29.99, then annual membership by level: Student £49.99, Associate £99.99, Professional £199.99, Fellow £299.99. You hold the credential through continuing professional development rather than re-sitting an exam every few years.

What makes this credential worth more than another certificate?+

Because the Institute of AI is the UK's professional body for AI, and the standard is published for you to read before you apply. Every level is judged against that competency framework, held live by an annual CPD target, and verifiable publicly, so an employer is trusting the standard rather than your account of yourself.

Assessed, not assumed