Artificial Intelligence

How to use this topic

The first seven pages build the foundations: what AI is, how machines learn from data, the classic algorithms, how to evaluate a model, and how neural networks and deep learning work. The next five pages are about language models and generative AI — transformers, working with language models, embeddings and retrieval, agents, and vision and speech. The last seven pages are about AI in practice: in IT operations, the infrastructure behind it, security, responsible use, regulation, its limits, and using AI at work.

The Python examples use only the standard library, so you can see every step of the calculation; real projects use libraries such as scikit-learn or PyTorch that do the same work faster. Every example was run and prints exactly the output shown. AI changes quickly: product names, model versions and legal deadlines are dated on the pages, and the principles are what lasts.

Pages are tagged with the certifications they help with: AI-901 (Microsoft Certified: Azure AI Fundamentals, the successor of AI-900) and AWS AI Practitioner (AWS Certified AI Practitioner, exam AIF-C01).

Learning path

1. 01What Is AI?What artificial intelligence means and where it came from: definitions, narrow and general AI, the main approaches, and the history from Turing and Dartmouth through the AI winters to deep learning and today's language models.BeginnerAI-901AWS AI Practitioner
2. 02How Machines LearnThe core idea of machine learning: learning patterns from data instead of writing rules, features and labels, training and inference, and supervised, unsupervised and reinforcement learning.BeginnerAI-901AWS AI Practitioner
3. 03Data for AIWhy data decides how good a model can be: collecting and labeling data, data quality, splitting into training, validation and test sets, data leakage, and bias that is already in the data.BeginnerAWS AI Practitioner
4. 04Classic Machine LearningThe algorithms that still do most practical work: linear and logistic regression, decision trees, k-nearest neighbors and k-means clustering, each built from scratch in a few lines of Python.IntermediateAI-901AWS AI Practitioner
5. 05Evaluating ModelsHow to tell whether a model is any good: the confusion matrix, accuracy, precision, recall and F1, overfitting and underfitting, cross-validation, and choosing the metric that matches the cost of mistakes.IntermediateAWS AI Practitioner
6. 06Neural NetworksHow a neural network turns numbers into predictions: neurons and weights, layers and activation functions, the loss function, and learning by gradient descent and backpropagation, shown step by step in Python.IntermediateAWS AI Practitioner
7. 07Deep LearningWhat changes when networks get deep: convolutional and recurrent networks, training on huge datasets with GPUs, transfer learning and fine-tuning, and why deep learning took off after 2012.IntermediateAWS AI Practitioner
8. 08Transformers and Large Language ModelsHow language models work: tokens and embeddings, the transformer and attention, pretraining on text, fine-tuning and learning from human feedback, the context window, and reasoning models.IntermediateAI-901AWS AI Practitioner
9. 09Working with LLMsGetting useful results from language models: prompts and system prompts, giving context and examples, temperature, tokens and cost, structured output, and calling a model through an API.BeginnerAI-901AWS AI Practitioner
10. 10Embeddings and RetrievalRepresenting meaning as numbers: embeddings and vector similarity, semantic search and vector databases, and retrieval-augmented generation (RAG) to ground answers in your own documents.IntermediateAI-901AWS AI Practitioner
11. 11AI Agents and Tool UseWhen a model does more than answer: tool calling, the agent loop of planning, acting and observing, the Model Context Protocol, multi-agent setups, and the guardrails that keep agents safe.IntermediateAI-901AWS AI Practitioner
12. 12Vision, Speech and Multimodal AIAI beyond text: image classification and object detection, optical character recognition, speech recognition and synthesis, multimodal models, and generating images, audio and video with diffusion models.IntermediateAI-901AWS AI Practitioner
13. 13AI in IT OperationsHow AI is used in IT teams today: AIOps for monitoring and incidents, code and configuration assistants, AI on the service desk and in network operations, and where human judgment stays essential.Intermediate
14. 14AI Infrastructure and Cloud ServicesWhat it takes to run AI: GPUs, TPUs and NPUs, training versus inference costs, cloud AI services from Microsoft, Amazon and Google, open-weight models, running models locally, and energy use.IntermediateAI-901AWS AI Practitioner
15. 15AI SecurityHow AI systems are attacked and defended: prompt injection and jailbreaks, data poisoning, model theft and data leakage, the OWASP Top 10 for LLM applications, and securing agents and AI-assisted work.IntermediateAWS AI Practitioner
16. 16Responsible AIBuilding and using AI that people can trust: fairness and bias, transparency and explainability, privacy, human oversight, accountability, environmental impact, and lessons from real failures.BeginnerAI-901AWS AI Practitioner
17. 17AI Regulation and GovernanceThe rules around AI: the EU AI Act and its risk levels and deadlines, AI law in the US and the UK, how data protection law applies to AI, governance standards and frameworks such as ISO/IEC 42001 and the NIST AI RMF, and who supervises AI.IntermediateAWS AI Practitioner
18. 18Limits and MythsWhat today's AI can and cannot do: hallucinations, brittle reasoning, missing world knowledge, benchmark limits, the debate about artificial general intelligence, and how to read AI claims critically.Beginner
19. 19Using AI at WorkPractical habits for using AI tools responsibly in an IT job: what to share and what not, checking output, keeping a human accountable, organizational AI policies, AI literacy, and certifications to consider.Beginner

Artificial Intelligence glossary

Short definitions of the terms used in this topic. The full glossary covers every topic.