This list is for the foundations rather than any one language: the theory, the math, and the core concepts that sit underneath everything else. Some of these channels are full university courses; others are the kind of visual explainers that finally make an idea click. Mix the two.
Full courses and lectures
Harvard’s CS50 — If you start anywhere, start here. Harvard’s legendary intro to computer science, plus spin-offs on AI with Python, web programming, and game development. Rigorous but genuinely welcoming to beginners.
MIT OpenCourseWare — Real MIT lectures across math, computer science, and engineering. Dense, authoritative, and completely free.
stanfordonline — Stanford course material spanning security, NLP with deep learning, and engineering leadership. Best once you have the basics and want depth.
Concept explainers and visual intuition
3Blue1Brown — The gold standard for visual math. The “Essence of Calculus” and “Essence of Linear Algebra” series, plus a superb intro to neural networks. Watch these before any ML course.
Computerphile — Short, expert-led explainers on how computers really work: memory, algorithms, AI, and security. Perfect for filling a specific gap in an afternoon.
Domain of Science — Big-picture “maps” of entire fields, from mathematics to quantum physics. Great for seeing how the pieces of a subject fit together.
Zach Star — Applied math and the different engineering disciplines, framed around where each one actually gets used. Useful if you are still deciding what to specialise in.
More theory and math
Neso Academy — Structured, exam-style courses on networks, digital electronics, and signals and systems.
Bob Trenwith and FCIH OCW — Two deep archives of mathematics, computing, and data-science playlists for when you want to drill a specific topic.
If you want one path: watch a few 3Blue1Brown videos to build intuition, then commit to CS50 for the structure.