Courses, Books and Resources
A short, curated list of well-established ways to go further — sorted by level, marked free or paid, with Malaysian programmes included.
Search for “learn AI” and you will find thousands of courses, many of them repackaged hype with a certificate attached. The good news is that some of the best material in the world is free, written by the people who built the field, and has been tested by millions of learners. This page lists only well-established options. It is deliberately short: one good course finished beats five started.
How to use this list
- Match the level. The tabs below follow the five tracks of this site. If a course feels impossible, drop a level; if it feels trivial, move up.
- Free means free to learn. Several courses are free to watch or read but charge for a certificate. We mark them Free, Free + paid certificate or Paid. Prices and terms change often, so check the provider’s page As of October 2026.
- Prefer doing over watching. The courses with exercises, notebooks or projects are worth more than lecture-only ones.
- Certificates are optional. For most jobs, a small project you can show and explain says more than a badge.
Courses by level
- Elements of AI — University of Helsinki with MinnaLearn. A gentle, text-based introduction to what AI is and what it is not, no maths or code. Available in many languages. Free.
- AI For Everyone — DeepLearning.AI on Coursera, taught by Andrew Ng. Non-technical: what AI can do, how projects work, and its effect on organisations and society. Free to learn on Coursera’s usual terms + paid certificate.
- AI untuk Rakyat — see the Malaysian section below. Free.
- This site’s Novice track — twelve short pages. Free.
- Google AI Essentials — Google, delivered through Coursera (launched 2024). Practical use of generative AI at work: prompting, productivity and responsible use. Paid (a modest fee; financial aid available) As of 2026.
- DeepLearning.AI short courses — one- to two-hour courses, many made with AI companies, on prompting, agents and specific tools. Some assume a little Python. Free to watch; quizzes, certificates and some hands-on features need a paid membership As of 2026.
- The official help centres and prompting guides of the assistant you use (ChatGPT, Claude, Gemini, Copilot and others) — dry but accurate and up to date. Free.
- This site’s Practitioner track. Free.
- CS50’s Introduction to Artificial Intelligence with Python — Harvard. Search, knowledge, uncertainty, optimisation, learning, neural networks and language, with weekly Python projects. Assumes some Python. Free + paid certificate (via edX).
- Machine Learning Specialization — DeepLearning.AI and Stanford Online, taught by Andrew Ng (2022). Regression, classification, neural networks, trees, clustering and recommender systems in Python. Coursera terms; paid certificate.
- Practical Deep Learning for Coders — fast.ai (Jeremy Howard). Top-down: train working models first, understand them afterwards. Free.
- Hugging Face courses — the LLM Course, Agents Course and others, built around open models and the Hugging Face libraries. Free.
- Kaggle Learn — short browser-based micro-courses in Python, pandas and machine learning. Free.
- Neural Networks: Zero to Hero — Andrej Karpathy. Videos that build backpropagation, language models and a GPT from scratch in code. Free.
- MIT 6.S191: Introduction to Deep Learning — MIT. A fast annual bootcamp with lectures and labs, refreshed each year. Free.
- Stanford CS229: Machine Learning — the mathematical treatment of classic machine learning; lecture notes online and past lectures on YouTube. Free (materials).
- Stanford CS231n: Deep Learning for Computer Vision — CNNs and modern vision; famous assignments. Free (materials).
- Stanford CS224n: Natural Language Processing with Deep Learning — from word vectors to Transformers and LLMs. Free (materials).
- Dive into Deep Learning — interactive textbook with runnable code. Free.
- Sutton & Barto, Reinforcement Learning: An Introduction — the standard text on reinforcement learning, 2nd edition (MIT Press, 2018). Free online; paid print.
- arXiv and OpenReview — where new papers and their public reviews appear. Free.
- Conference talks and tutorials from NeurIPS, ICML, ICLR, ACL and CVPR — many are recorded and published online. Mostly free.
- The Stanford courses in the Advanced tab, read alongside their cited papers, and this site’s Expert’s Road. Free.
Books worth owning
Tap a spine. Free online editions are noted; the rest are paid but often in libraries.
Malaysian programmes
Malaysia has run several national initiatives to raise AI literacy. Details change often, so check the official sources before enrolling As of October 2026.
- AI untuk Rakyat (“AI for the People”) — launched in early 2024 by MyDIGITAL Corporation with Intel as a free, self-paced online course of roughly four hours in two modules (AI Aware and AI Appreciate), available in Bahasa Melayu, English, Mandarin and Tamil. A revised six-module programme for Malaysians aged 18–30 was announced in August 2026 As of 2026. Free.
- AI Malaysia (formerly the National AI Office, NAIO) — the national AI agency under the Ministry of Digital, which coordinates AI policy and lists learning resources and events; it institutionalised NAIO in July 2026 As of 2026. Free to browse.
- MDEC (Malaysia Digital Economy Corporation) — runs and supports digital-skills and AI training initiatives for students, workers and companies, often with industry partners. Offerings vary by year. Many free or subsidised.
For more on the country’s policy and industry, see AI in Malaysia.
Choosing well
Should I pay for a certificate?
Only if an employer or institution asks for it, or if paying helps you commit. The learning is usually identical in the free version.
How do I spot a low-quality AI course?
Be wary of promises of guaranteed jobs or income, courses that are mostly lists of tools, no named instructors with real track records, and pressure to buy quickly. The courses above are run by universities, established companies or researchers with published work.
Video course or book?
Whichever you will actually finish. Videos are good for intuition; books and notebooks are better for depth and reference. Many people pair one of each.
I started a course and got stuck. Now what?
Drop down a level for a week, or find the exact concept you are missing on this site or in the glossary, then go back. Getting stuck is normal; quitting is optional.
// Pause and reflect
- Which single course or book from this page will you start this week — and which will you deliberately leave for later?
- Do you learn better from videos, text or hands-on exercises? How should that change your choice?
Your notes are saved privately in this browser.
Key terms on this page
Sources & further reading
- University of Helsinki & MinnaLearn, Elements of AI (2018 onwards).
- Coursera, “Google launches AI Essentials course on Coursera” (2024).
- Lowyat.NET, “Govt launches AI Untuk Rakyat program” (2024).
- Malay Mail, “AI for the People participants must complete six modules to receive free AI tools” (2026).
- Richard S. Sutton & Andrew G. Barto, Reinforcement Learning: An Introduction, 2nd edition (MIT Press, 2018).
- Stuart Russell & Peter Norvig, Artificial Intelligence: A Modern Approach, 4th edition (Pearson, 2020).