<Generics>

Kafka Basics

Advanced Kafka

Kafka

Test your understanding of AI bias, the black-box problem, privacy risks, and responsible AI development.

AI Ethics and Limitations

Test your understanding of the AI development pipeline from problem definition to production deployment.

How AI Systems Are Built

Test your understanding of AI fundamentals, history, and the difference between narrow and general AI.

What is Artificial Intelligence?

Test your understanding of what deep learning is, what it excels at, and why the field has taken off.

Introduction to Deep Learning

Test your understanding of logistic regression — what it classifies, how sigmoid works, and how variables are defined.

Logistic Regression

Test your understanding of gradient descent — the update rule, learning rate, gradient formulas, and batch variants.

Gradient Descent for Logistic Regression

Test your understanding of overfitting, underfitting, bias, variance, and how to spot each from training and test error.

Overfitting in Logistic Regression

Test your understanding of the three approaches to fix overfitting, the regularized cost function, and the lambda hyperparameter.

Regularization for Logistic Regression

Test your understanding of the regularized cost function, its gradient, and the weight decay update rule.

Regularized Logistic Regression

Test your understanding of how artificial neurons work, what ReLU does, how hidden layers learn, and why neural networks are a form of supervised learning.

Neural Networks Explained

A quick check on real-world neural network applications and the four main network types.

Neural Networks in Practice

Test your knowledge of the Transformer architecture, attention mechanisms, and modern LLMs.

Transformers & Attention

Test your understanding of supervised, unsupervised, and reinforcement learning — and where deep learning and neural networks fit in.

Types of Machine Learning

Test your understanding of underfitting, overfitting, and how to diagnose and fix each type of model error.

Bias-Variance Tradeoff

Test your understanding of classification — fixed output categories, binary vs multi-class problems, and how classification differs from regression.

Classification

Test your understanding of precision, recall, F1 score, ROC-AUC, and when to use each metric.

Classification Metrics

Test whether you can tell classification problems from regression problems at a glance.

Classification vs Regression

Test your understanding of why feature scaling matters for gradient descent and how to apply max, mean, and z-score normalisation.

Feature Engineering

Test your understanding of the gradient descent algorithm — update rules, convergence, the cost surface, and why simultaneous updates matter.

Gradient Descent

Test your knowledge of what machine learning is, who coined the term, and the three core learning paradigms.

Intro to Machine Learning

Test your understanding of the learning rate α — what it controls, what goes wrong when it is too large or too small, and why gradient descent stops automatically at the minimum.

The Learning Rate

Test your understanding of the linear regression model — the f(x) = wx + b formula, parameters w and b, training notation, and making predictions.

Linear Regression

Test your understanding of multivariate linear regression — how the model extends to multiple features, what changes in the gradient descent update rules, and why simultaneous updates matter.

Regression with Multiple Inputs

Test your understanding of overfitting, underfitting, the bias-variance naming, and regularization concepts.

Overfitting & Regularization

Test your understanding of regression — predicting continuous values, the crop yield problem, and how regression differs from classification.

Regression

Test your understanding of the regularized cost function, weight decay, and how the gradient changes when regularization is added.

Regularized Linear Regression

Test your understanding of how supervised learning works — labeled data, the input-output pattern, training/validation/test splits, and regression vs classification.

Supervised Learning

Test your understanding of unsupervised learning — what it is, how clustering works, anomaly detection, and dimensionality reduction.

Unsupervised Learning

Test your understanding of NumPy broadcasting rules and how arrays of different shapes can be combined.

Broadcasting in Python

Test your knowledge of Jupyter Notebooks — cell types, shortcuts, and best practices for ML work.

Jupyter Notebooks for ML

Check your understanding of how ML datasets are structured as matrices, what shapes mean, and how to reshape arrays.

Matrices and Data Representation

Test your knowledge of core NumPy operations — array creation, indexing, axis reductions, and matrix math.

NumPy Fundamentals

Test your understanding of why Python is the language of machine learning and the basics you need to get started.

Python for AI & ML

Test your knowledge of the scikit-learn API — fit/predict, train/test split, scaling, and evaluation.

Scikit-learn Essentials

Test your understanding of why vectorization matters and how to replace loops with fast NumPy array operations.

Vectorization in Python

Discover what deep learning is, where it is already changing the world, what it is best suited for, and why the field has exploded in the last decade.

Learn what machine learning is, why it matters, and how the three main learning paradigms — supervised, unsupervised, and reinforcement learning — work.

Introduction to Machine Learning

From a single neuron to deep networks — understand how neural networks work through a concrete crop yield prediction example.

Understand how supervised learning works — labeled training data, the input-output pattern, and the two types: regression and classification.

Understand the three learning paradigms — supervised, unsupervised, and reinforcement learning — and see why neural networks and deep learning are built around supervised learning.

See how neural networks power supervised learning at scale — from our crop yield example to five real-world applications used by the biggest companies in the world.

Neural Networks and Machine Learning in Practice

Learn what regression is, why it is used to predict continuous values, and how the crop yield prediction problem illustrates the core idea.

Learn what classification is, why the output is always a label from a fixed set of categories, and how real-world problems like spam detection and fraud detection work.

Implement logistic regression from scratch — variables, sigmoid activation, loss function, and a complete binary classifier.

A concise side-by-side comparison of classification and regression — output types, what each asks, and how to instantly tell them apart.

Implement the full gradient descent training loop for logistic regression — gradients, update rule, and evaluation.

Understand the attention mechanism, encoder-decoder structure, positional encoding, and why Transformers power modern LLMs like GPT and BERT.

Transformer Architecture

Learn what unsupervised learning is, why it needs no labels, and how its three main types — clustering, anomaly detection, and dimensionality reduction — work.

Learn why accuracy alone misleads, and how precision, recall, F1 score, and ROC-AUC give a complete picture of classifier performance.

Understand why models underfit or overfit, how to diagnose each using training and test error, and how model complexity drives the fundamental tension between bias and variance.

The Bias-Variance Tradeoff

A clear, jargon-free introduction to AI — what it means, how it evolved, the difference between narrow and general AI, and why it matters now.

Follow an AI system from raw idea to deployed product — data collection, model training, evaluation, and the engineering behind production AI.

Understand overfitting and underfitting through a crop yield regression example, then fix it with three practical approaches — ending with the regularized cost function and the lambda hyperparameter.

Where AI fails, how bias enters models, the black-box problem, privacy risks, and what responsible AI development looks like in practice.

A practical introduction to Python as the language of machine learning — what you need to know and why it matters.

Learn how ML datasets are structured as matrices, how to read shapes, and how to reshape arrays for model input.

Understand what a cost function is, why it is essential for training, and how the squared error cost function J(w,b) measures how well a linear model fits the data.

The Cost Function

Understand how gradient descent finds the optimal w and b by repeatedly moving downhill on the cost surface — including the formal algorithm, a 3D visualisation, a Python implementation, and the role of the learning rate.

Understand what the learning rate α is, how it controls every gradient descent step, and what happens when you choose a value that is too large, too small, or just right.

Build your first supervised learning model step by step — the crop yield prediction problem, ML notation, and the linear function f(x) = wx + b with an interactive explorer.

Understand why vectorization replaces loops in ML code and how to write fast, array-based operations with NumPy.

Learn NumPy's broadcasting rules for operating on arrays of different shapes without loops or manual tiling.

Master the NumPy array operations that appear in every ML codebase — creation, indexing, math, and matrix operations.

Learn the fit/predict API, train/test splitting, feature scaling, and evaluation metrics that every ML practitioner uses daily.

Learn why features with very different ranges slow gradient descent, and how to fix it with max normalisation, mean normalisation, and z-score normalisation.

Set up and use Jupyter Notebooks effectively — the interactive coding environment used by every ML practitioner.

Extend linear regression and gradient descent to multiple input features — add rainfall, temperature, and soil quality to the crop yield model and implement the full multi-parameter update loop in Python.

Build a two-layer neural network from scratch in pure NumPy — neurons, layers, forward propagation, and loss functions.

Basics of Neural Networks

Understand overfitting and underfitting through crop yield regression and diabetic classification examples — with interactive diagrams showing high bias, good fit, and high variance.

Fix overfitting with regularization — understand the three approaches to reduce variance, then derive the regularized cost function and the lambda hyperparameter.

Derive the complete regularized logistic regression model — from the forward pass through to the regularized cost function and weight-decay gradient descent update rules.

Pick up from the regularized cost and derive what changes — the new gradient for each weight, the weight-decay update rule, and a full Python implementation.

AAI

Courses

Explore Quizzes

🚀 REST API Missions

🚀 GET the Data

📤 POST your Passport

⚔️ Battle CRUDzilla

Why You'll Love This:

🔨 Challenges

Your Learning Progress