Loading
Loading

Technology & AI courses
Intermediate levelTake this machine learning course online and build models in Python that hold up on data they've never seen. Following a fictional UK homeware retailer, you'll frame the problem, catch the leaks that flatter a score, build scikit-learn pipelines, set decision thresholds from what each mistake costs, check a model for fairness and ship one as a live demo. The maths stops at GCSE level.
Course overview
This course takes you from a first decision tree in your browser to a model you can defend at interview. You'll frame problems and explore data, split and test without leakage and prepare features in pipelines, then work through regression and forecasts, classification and metrics, trees and ensembles, unsupervised learning and neural networks, before explaining models, checking fairness, putting one to work and finishing with a capstone of your choice.
By the end you can turn a business question into a prediction task, or show it doesn't need one, train and tune models that hold up on new data, and explain every choice to an interviewer or a regulator. It suits analysts, developers and Python learners ready to predict rather than report.
Free course guide
Every module and what it covers, the price, how you're assessed and where the course can take you.
10-page PDF, free
Curriculum
An indicative outline of the modules and what each covers. Work through them in order, at your own pace.
Key benefits
Everything you need to learn the skill, prove it and put it to work.
Getting started
There is no application, no waiting for a start date and no sales call in the way.
Pay once. Instant access, yours for life.
Career outcomes
The request to be ready for in a data science interview is 'talk me through a model you built'. You finish with a certificate employers can verify and a GitHub portfolio, with a live app in it, that answers it.
Source: National Careers Service (gov.uk), September 2026.
An example of the certificate for the Certificate in Machine Learning Fundamentals.


If you're new to Python, start with Python Programming, then Data Analysis with Python. If you want statistics and the data science workflow first, Data Science Fundamentals is the gentler start. And if you'll be judging AI proposals rather than building models, AI for Business Leaders fits better.
A course advisor will talk it through with you free, and will say so if a different course fits your goal better.
Good to know
It follows a fictional UK homeware retailer from a first returns model to a costed churn model, a review classifier and a fairness check. You start by training a quick model, then learn to frame problems, split and test without leakage and build feature pipelines. Later modules cover regression and forecasts, classification and choosing metrics, trees and ensembles, finding patterns without labels, neural networks and text, explaining models and checking fairness, and putting a model to work as an API and a demo app. A capstone of your choice closes it.
Yes. This is a hands-on Python course: you'll write the code, build pipelines in scikit-learn and ship a model as an app. You need basic Python and everyday pandas before you start, and Python Programming and Data Analysis with Python are the courses that get you there.
No more than GCSE level, and no statistics background is assumed. The course explains the ideas it needs as it goes, and most of the effort goes into judgement: what to predict, how to test it and what a mistake costs.
No. Python, scikit-learn and the other libraries are free and open source, and most of the course runs in Google Colab's free tier. The language model API used in the text lessons is optional.
Data Science Fundamentals if you're new to code and want statistics and the analysis workflow first. Machine Learning Fundamentals once you can use Python and pandas and want to build models that predict. Taken in that order, the first sets up the second.
You don't need data from work. The lessons follow a fictional homeware retailer, and the capstone briefs use real UK open data, such as DVSA MOT results and Companies House filings. A problem from your own work is welcome if your employer has cleared the data.
It gives you what an interviewer will ask about: a model you built, tested and shipped, and the judgement to explain every choice in it. Machine learning roles usually ask for strong programming and statistics as well, so think of it as a strong step on that path.
Yes. Pass the final assessment and you get an OBA certificate of completion that an employer can verify online. It isn't a cloud provider's machine learning certification. Your capstone and live demo are the other half of the proof.
Download the free info pack first. It walks you through every module, what you'll build and what you need before you start, so you can enrol knowing exactly what you're getting. If you'd like to talk it through, get in touch and we'll help you choose.
Keep exploring
Intermediate
Beginner
Beginner
Beginner
Free info pack
Every module, the price, how you're assessed and where it can take you. Yours free, and it opens straight away.
Takes 30 seconds