Learn data science using
data that actually matters.
Most data science courses use American housing prices or global e-commerce data. Code Caribbean uses Jamaican crime statistics, Caribbean climate datasets, local election results, and CARICOM trade records.
When the data is familiar, the learning sticks, and that is not just a nice idea. New knowledge anchors to what you already know, so a Kingston rainfall dataset hooks onto the weather you have lived your whole life, while abstract foreign data has nothing to grab. You understand why a model is underfitting because you know what the actual weather in Kingston looks like. You spot outliers because you know Jamaican agriculture.
This is a project-driven program, and that is a deliberate choice about how people actually learn. You do not spend months in theory before touching real problems, because skills grow when you use them, not when you store them for later. From week one, you are cleaning, exploring, and building with real Caribbean data, and the curiosity that comes from working on a question you genuinely care about is the engine that keeps you coding when the bug will not budge.
Caribbean Datasets Throughout
Every dataset, every project, every case study uses regional data. You graduate knowing how Caribbean data behaves.
Project Portfolio Included
Five completed projects on GitHub by the end of the program. Employers see real work, not just a certificate. Building five times, not once, is spaced practice: each project pulls forward skills from the last, so the early lessons get recalled and strengthened instead of fading.
🧠 Why learning on Caribbean data works
Two well-established ideas about learning sit underneath this whole programme. The first is that we understand new things by connecting them to things we already know. Familiar local data, the weather you grew up in, the crops you have seen at market, gives every new concept somewhere to land, which is why a regression on Kingston rainfall makes more sense faster than the same maths on an abstract foreign dataset.
The second is that skill is built by retrieval, not recognition. Re-reading a notebook until it "looks familiar" feels productive but barely moves you; pulling the answer out of your own head, writing the code yourself, getting it wrong, and fixing it is what actually wires it in. Code Caribbean is structured around that loop. Instructors scaffold the hard parts early, giving you a worked example to lean on, then deliberately fade the support so that by the capstone you are the one doing the work that used to need a hand.
Prior knowledgeRetrieval practiceScaffoldingIntrinsic motivation
