Our educational approach
Good teaching makes
the thinking visible.
You should be able to see what you're learning, why it matters and how you will know you've understood it.
Start with what learners need to do
Each module starts with something a learner should be able to do. We then agree the evidence that would show it, choose a practice task and explain the ideas needed to finish it. A list of tools is not a curriculum.
- Start hereOutcomeWhat should you be able to do?
- Make it observableEvidenceWhat would show understanding?
- Build confidencePracticeTry, discuss and improve.
- Carry it forwardApplicationUse the skill with appropriate support.
Practice with feedback
18 core modules progress from Beginner to Intermediate and Advanced. Each core module has three taught sections that connect an explanation, a worked example, guided practice and a task to try on your own. Public, student and business applications give learners a choice of contexts that feel relevant. Eight AI Edge extensions add model foundations, current developments, evidence tracing, data, multimodal AI, benchmarks, infrastructure and startup evaluation.
Three practice quizzes contain 54 scenario questions. Every answer option comes with an explanation, so a mistake points you straight to the reasoning to revisit. 27 teaching graphics and 26 short think-and-check activities, fictional source packs, a workbook and two local labs support the written teaching.
Self-paced study includes written practice feedback. Tutor-led sessions and individual feedback are arranged separately, and we'll confirm availability and support terms with you beforehand.
Assess understanding, not polish
The practice projects combine a demonstration, a test log and an explanation of your decisions. They ask learners to protect information, check claims and handle unfamiliar or failing examples.
AI assistance is part of the practical work, and you should say where you used it. It can't replace your own explanation. A practice score, attendance or a completion tick on its own doesn't show real competence.
If you'd like your work assessed by a person, we'll agree a fresh task, a named reviewer with the right experience and clear criteria. This is arranged separately from self-paced enrolment. The course does not offer a regulated qualification or academic credit.
Make learning approachable
The public sample uses plain language, text alongside its visual example, keyboard-accessible controls and no timer. Try it without an account. Enrolled learners can revisit the lessons and download the PDF, structured HTML edition and practice files.
Beginner study needs no coding experience. Advanced local labs need basic file skills and Python setup; they include installation checks and worked instructions. Tool practice asks you to check the features available in your own account before using them.
Tell us about access or learning-support needs before enrolment. Agree formats, equipment and adjustments with us; do not assume everyone has the same device, confidence or paid software. Native portal controls have limitations, and the course does not claim formal accessibility certification.
Keep people in the conversation
Questions about mistakes, job changes and responsible use deserve a thoughtful answer. The exercises make those decisions visible. Where we arrange tutor support, the tutor should explain the limits of an exercise. They should not promise that a tool will remove uncertainty or replace professional judgement.
Improve the course with evidence
The October edition passed two quality sweeps covering content, source checks, practice questions, local lab behaviour and the published learner pages. These checks establish what was inspected; they do not establish learner satisfaction or workplace results.
Tool instructions should be revised when behaviour changes. Any reported result should say what was measured, over what period and with which limitations. Learner pilot results, study-time calibration and assessment moderation remain work for delivery. Before tutor-led delivery, we also need named staff and agreed support and complaints arrangements.
The current learner edition was reviewed on 3 October 2026 and lives in our private learning space. Mentioning national frameworks does not make Ampliflow accredited, government approved or a higher-education awarding body.
Reference points, not endorsements
Built with the wider
education community in mind.
We use these public sources to inform the design. The learning exercises are our own; the organisations below have not approved or endorsed this programme.
- Skills England: AI foundation skills for work
A reference for the foundational skills in our curriculum.
- Education and Training Foundation: professional standards
A reference for reflective, inclusive teaching and professional development.
- QAA: UK Quality Code
Higher-education guidance that informs our approach to learning and assessment. This is not a higher-education award.
- Jisc: student perceptions of AI
Learner perspectives on clear guidance, practical support and the importance of human teaching.
- Skills for AI: what works for upskilling
Workplace learning research that informs practice and follow-up.
Source review: 28 September 2026. The UK Quality Code concerns higher education; our initial offer is a non-regulated workplace short course.