Learn the ideas. Put them to work.

Understand AI,
one idea at a time.

A practical path from the foundations to building with AI. Learn how models work, where they fall short, and how to check what they produce.

No machine-learning background required. Start at the beginning or choose the question you have today. The developer guides use small code examples and familiar application scenarios.

Start with AI foundations →

Looking for the AI Assistant? It has moved to /assistant. Sign in there for coding help.

Stage 1

AI Fundamentals

Build a clear mental model of learning, prediction and generation.

  1. Artificial Intelligence vs Machine Learning vs Deep Learning

    Understand the relationship between AI, machine learning and deep learning, and choose an appropriate approach for a real task.

    BeginnerAbout 8 min readPractical exerciseNo prior AI knowledge needed

  2. How AI Models Learn: Training and Inference

    Follow data through training, evaluation and inference, and learn why a model can perform well in training yet fail in use.

    BeginnerAbout 6 min readPractical exerciseBefore you start: AI, machine learning and deep learning

  3. What Is Generative AI?

    Learn what generative models produce, where they help, and how to evaluate outputs without confusing fluency with accuracy.

    BeginnerAbout 5 min readPractical exerciseBefore you start: Training and inference

  4. How Large Language Models Work

    Explore tokens, learned representations, attention and next-token generation through a practical language-model workflow.

    BeginnerAbout 5 min readPractical exerciseBefore you start: Generative AI

  5. Tokens and Context Windows

    Plan a context budget, understand tokenization, and manage long conversations without silently losing important information.

    BeginnerAbout 5 min readPractical exerciseBefore you start: Large language models

  6. What Are Embeddings?

    Understand vector representations and semantic retrieval, including the limits of similarity scores and document search.

    IntermediateAbout 5 min readPractical exerciseBefore you start: Large language models · Basic JavaScript for the example

  7. Why AI Hallucinates

    Recognize unsupported claims and invented APIs, then use evidence, evaluation and abstention to reduce their impact.

    BeginnerAbout 5 min readPractical exerciseBefore you start: Generative AI
Stage 2

Using AI Tools

Use everyday AI tools with clear instructions and independent checks.

  1. Using ChatGPT

    Understand the chat workflow, give useful context, and check answers and tool results against evidence.

    BeginnerAbout 8 min readPractical exerciseBefore you start: Generative AI

  2. Using Claude

    Work with documents and coding tasks in Claude, while checking claims, permissions and the results of any tools.

    BeginnerAbout 8 min readPractical exerciseBefore you start: Generative AI

  3. Using Google Gemini

    Use Gemini for a bounded task, understand grounding and connected tools, and distinguish a generated answer from verified evidence.

    BeginnerAbout 8 min readPractical exerciseBefore you start: Generative AI

  4. Prompt Engineering for Developers

    Write clear tasks, supply relevant context and test prompt changes against examples instead of relying on magic phrases.

    BeginnerAbout 6 min readPractical exerciseBefore you start: Tokens and context windows

  5. Using AI for Research

    Turn a research question into a source plan, verify citations and build conclusions that reflect the available evidence.

    BeginnerAbout 8 min readPractical exerciseBefore you start: Hallucinations and reliability

  6. Using AI for Programming

    Specify a small programming task, inspect the proposed implementation and test it against independent examples.

    IntermediateAbout 10 min readPractical exerciseBefore you start: Basic JavaScript · Prompt engineering

  7. AI Agents and Tool Use

    Follow an agent loop and learn how tool contracts, permissions, confirmation and stopping rules constrain its actions.

    IntermediateAbout 10 min readPractical exerciseBefore you start: Prompt engineering · Basic JavaScript
Stage 3

Building with AI

Build reliable applications around models, data and controlled actions.

  1. Working with AI APIs

    Design a server-side AI integration with validation, timeouts, cost limits and useful handling of incomplete or invalid results.

    IntermediateAbout 7 min readPractical exerciseBefore you start: HTTP requests · Basic server-side programming

  2. Integrating AI into a Website

    Connect a browser interface to a controlled server boundary, with clear loading, cancellation and failure behavior.

    IntermediateAbout 8 min readPractical exerciseBefore you start: AI APIs · JavaScript fetch and async/await

  3. Building an AI Chatbot

    Manage conversation state, context budgets and concurrent messages without treating generated replies as trusted facts.

    IntermediateAbout 7 min readPractical exerciseBefore you start: AI APIs · JavaScript arrays and functions

  4. Structured Outputs and Function Calling

    Distinguish schema-constrained data from proposed tool actions, then validate both structure and authority before use.

    IntermediateAbout 8 min readPractical exerciseBefore you start: AI APIs · JSON and basic validation

  5. Retrieval-Augmented Generation

    Build an evidence path from documents to answers, with access controls, useful chunks, citations and retrieval evaluation.

    IntermediateAbout 8 min readPractical exerciseBefore you start: Embeddings · AI APIs

  6. Vector Databases and Similarity Search

    Understand vector indexes, distance metrics and filtered retrieval, and evaluate search quality without confusing similarity with truth.

    IntermediateAbout 9 min readPractical exerciseBefore you start: Embeddings · Database basics

  7. How to Verify AI-Generated Code

    Review generated code against requirements, independent tests, security boundaries and the dependencies you actually use.

    IntermediateAbout 6 min readPractical exerciseBefore you start: Basic JavaScript and tests

  8. Using AI to Debug Code

    Turn a reproducible failure into a focused investigation, test hypotheses and verify the smallest justified correction.

    IntermediateAbout 16 min readPractical exerciseBefore you start: JavaScript functions, promises and async/await

  9. Building AI Features Securely

    Separate model suggestions from authority, protect private data and constrain tools before adding AI to an application.

    IntermediateAbout 7 min readPractical exerciseBefore you start: AI APIs · Authentication and authorization basics
Stage 4

Generative & Multimodal AI

Explore how AI generates and interprets images, video, speech and other inputs.

  1. AI Image Generation

    Write an image brief, iterate on controlled changes and evaluate generated visuals for usefulness, consistency and disclosure.

    BeginnerAbout 8 min readPractical exerciseBefore you start: Generative AI

  2. How Diffusion Models Work

    Explore noise, denoising and conditioning through a careful model of the diffusion process and its limitations.

    IntermediateAbout 9 min readPractical exerciseBefore you start: Training and inference · Generative AI

  3. AI Video Generation

    Plan a short generated scene and evaluate motion, temporal consistency, editing constraints and the final delivery.

    BeginnerAbout 8 min readPractical exerciseBefore you start: AI image generation

  4. AI Speech and Voice

    Distinguish transcription from speech generation, prepare audio tasks and evaluate intelligibility, errors and consent.

    BeginnerAbout 8 min readPractical exerciseBefore you start: Generative AI

  5. Multimodal AI

    Combine text, images and audio in a task while preserving evidence boundaries and testing what each input contributes.

    IntermediateAbout 9 min readPractical exerciseBefore you start: Large language models · Generative AI

  6. Computer Vision

    Distinguish classification, detection and segmentation, then evaluate predictions under realistic image conditions.

    BeginnerAbout 10 min readPractical exerciseBefore you start: Training and inference
Stage 5

Applied & Advanced AI

Apply AI in real systems, measure its results and examine its wider effects.

  1. AI and the Internet of Things

    Follow sensor data into a prediction, validate input quality and keep physical control decisions outside the model.

    IntermediateAbout 6 min readPractical exerciseBefore you start: Training and inference · Basic JavaScript for the example

  2. Edge AI

    Plan local inference around device memory, latency, energy and quality, including quantization and offline behavior.

    IntermediateAbout 5 min readPractical exerciseBefore you start: Training and inference

  3. AI Automation

    Place a bounded AI step in a reliable workflow with explicit state, human review, safe retries and recovery.

    IntermediateAbout 6 min readPractical exerciseBefore you start: AI agents and tool use · Basic JavaScript

  4. Fine-Tuning AI Models

    Decide when training is justified, prepare separate datasets and compare a fine-tuned model with simpler approaches.

    AdvancedAbout 6 min readPractical exerciseBefore you start: Training and inference · AI APIs

  5. Evaluating AI Models

    Build task-specific evaluations, interpret precision and recall, and examine failure cases before and after release.

    IntermediateAbout 6 min readPractical exerciseBefore you start: Training and inference · Basic JavaScript for the example

  6. AI Privacy, Safety and Ethics

    Map an AI feature’s data and effects, evaluate potential harms and define responsibilities, review and redress.

    BeginnerAbout 5 min readPractical exerciseBefore you start: Generative AI

Need a programming refresher? Browse the language documentation or experiment in the Playground.