Free · Built for African learners
A structured path from your first line of Python to models you trained yourself, on data about farms, clinics, markets and mobile money. It all runs in your browser, so any laptop will do.
How it works
No hour-long videos. Every idea is followed straight away by something you do, in real Python that runs in the page.
Learn a short idea, with a picture or a line of code you can run.
Play with an interactive until the idea clicks.
Build it in real Python, with instant checks and a mentor that reads your errors.
Explain it back in your own words, the real test of understanding.
Try it right here
Each dot is a maize farm: rainfall along the bottom, harvest up the side. Move the sliders until the line fits and watch the error shrink. That search for the best two numbers is exactly what “training” means. This is a real interactive from the AI & ML track.
yield = 0.030 × rainfall + 6.0
Average error (MAE)
3.30bags
The red lines are the errors — the gap between the line and each real farm.
A programme, not a playlist
Everyone starts with Python Essentials. Then choose a path, from the most approachable to the most demanding. Already code in Python? Take the placement check and go straight to step two.
STEP 1 · START HEREBeginner · no experience needed
Start from zero and learn the whole language: values, control flow, every built-in collection, functions in depth, generators, files, the standard library, object-oriented programming and professional habits. It's the foundation the Data Science, Data Engineering and AI & ML tracks all build on.
44 labs · 3 projects · 11 modules
PATH 1Intermediate
Difficulty 1 of 3Turn messy real-world data into decisions people trust.
Clean and join messy records, measure uncertainty honestly, run experiments, and work in the tools teams actually use — Excel, SQL, survey data and maps — ending with an investigation of real World Bank data.
30 labs · 5 projects · 7 modules
PATH 2Intermediate
Difficulty 2 of 3Pipelines, warehouses, and the systems that feed ML models.
Build the systems that move and model data: file formats and SQL, pipelines with Airflow-style DAGs and dbt, data quality and testing, warehouses, star schemas and streaming. It ends with an end-to-end mobile-money pipeline.
21 labs · 5 projects · 5 modules
PATH 3Advanced
Difficulty 3 of 3Train real models on real problems.
From your first straight-line model to neural networks, language models and responsible deployment — every concept built, run, and tested by you.
48 labs · 6 projects · 11 modules
The tools teams actually use, taught for real
Local data, real problems
Not another Titanic dataset. Labs use data about Kenyan farms, clinics and markets, mobile money, reviews in Swahili and English, and World Bank indicators for 20 African countries. Every module ends in a project for a realistic local client.
Python Essentials
Settle a savings group's arguments about who owes what, from payments typed into a phone.
Python Essentials
Model wallets, fees and transfers that never lose a shilling.
Data Science
Work out how much of a school programme's jump in exam scores it really caused.
Data Engineering
Build the weekly pipeline that tells a Kitale cooperative where its crops fetch the best price.
AI & Machine Learning
Tell a county agriculture office which maize fields to visit before blight spreads.
AI & Machine Learning
Fill a fraud team's daily review queue, with no fraud labels to learn from.
The World Bank indicators, census figures and county boundaries are real. The other datasets are illustrative, built to behave like the real thing.
Free · No card · No installs
Enroll in Python Essentials, open Lab 01, and run real code before you've finished your tea.
Start learning