Best free courses · Computer science
Best free university computer science courses on YouTube, ranked by total hours
Updated 2026-10-05 · 13 courses · totals are live
Harvard, Stanford, MIT and Berkeley put whole semesters of computer science on YouTube: the same lectures their students sit through, recorded in the lecture hall, free. Put together, the courses below come close to the core of a CS degree, from a first program to operating systems, networks and security, and that is the catch. A university course is built for a semester, so the length of the playlist is the first thing to know before you start one.
The numbers next to each course are live totals from the playlist itself, counted from the videos rather than from a syllabus. Click a course to see every lecture, tick off what you've watched, and get a finish date at your own pace.
From 5h 55m (CS50's Understanding Technology — Harvard) to 1d 19h 37m (MIT 6.001 Structure and Interpretation, 1986).
| # | Course | Videos | Total | At 1.5× |
|---|---|---|---|---|
| 1 | MIT 6.001 Structure and Interpretation, 1986 · MIT 6.001 Structure and Interpretation of Computer Programs (1986) Abelson and Sussman's SICP lectures in Lisp, recorded in 1986 for Hewlett-Packard engineers. Old video, lasting ideas: abstraction, recursion and how interpreters work. |
40 | 1d 19h 37m | 1d 5h 4m |
| 2 | CS 162: Operating Systems and Systems Programming - Berkeley · CS 162 Operating Systems and Systems Programming — Berkeley (Fall 2020) Processes and threads, synchronisation, scheduling, memory and file systems. Dense and long; the closest thing here to a real upper-division semester. |
27 | 1d 14h 49m | 1d 1h 52m |
| 3 | [CS161 FA23] Lectures · CS 161 Computer Security — Berkeley (Fall 2023) Memory-safety bugs, cryptography, web security and network attacks, from Berkeley's undergraduate security course, recorded in 2023. |
26 | 1d 10h 43m | 23h 9m |
| 4 | cs144 Introduction to Computer Networking · CS 144 Introduction to Computer Networking — Stanford How the internet works, layer by layer: packets, TCP/IP, routing and congestion control. Short videos, so it fits around a heavier course. |
145 | 1d 8h 33m | 21h 42m |
| 5 | CS50x 2025 Lectures · CS50x 2025 — Harvard Harvard's introduction to computer science: Scratch and C first, then Python, SQL and the web, with problem sets that are the real course. The best first course on this list, whatever you plan to do after it. |
13 | 1d 2h 22m | 17h 35m |
| 6 | MIT 6.824 Distributed Systems (Spring 2020) · MIT 6.824 Distributed Systems (Spring 2020) — Robert Morris Fault tolerance, replication and consistency, taught through the systems that made them famous: MapReduce, Raft, Spanner and more. Take it after operating systems. |
20 | 1d 1h 59m | 17h 19m |
| 7 | MIT 6.034 Artificial Intelligence, Fall 2010 · MIT 6.034 Artificial Intelligence (Fall 2010) — Patrick Winston Classic AI from before deep learning took over: search, constraints, rule systems and learning. Winston teaches how to think about intelligence, not how to call a library. |
30 | 1d 17m | 16h 12m |
| 8 | Course | Programming Methodology · CS106A Programming Methodology — Stanford (2008) Mehran Sahami's introductory programming course in Java, recorded in 2008. The tools have dated; the way it teaches decomposition and objects has not. |
28 | 22h 18m | 14h 52m |
| 9 | MIT 6.004 Computation Structures, Spring 2017 · MIT 6.004 Computation Structures (Spring 2017) — Chris Terman How a computer is built from the gates up: digital logic, a processor, caches and pipelining, in short topic videos. The hardware course software people skip and later wish they hadn't. |
172 | 19h 17m | 12h 52m |
| 10 | CS50's Web Programming with Python and JavaScript (CS50W) 2020 · CS50's Web Programming with Python and JavaScript (2020) — Harvard HTML and CSS, Git, Python, Django, SQL and JavaScript, building up to web apps you deploy. Picks up where CS50x's last weeks leave off. |
10 | 14h 13m | 9h 29m |
| 11 | CS50's Introduction to Artificial Intelligence with Python 2020 · CS50's Introduction to Artificial Intelligence with Python (2020) — Harvard Search, knowledge, uncertainty, optimisation, machine learning, neural networks and language, each with a Python project. The practical counterpart to 6.034. |
8 | 12h 40m | 8h 27m |
| 12 | Missing Semester IAP 2020 · The Missing Semester of Your CS Education — MIT (2020) The shell, Vim, Git, debugging and scripting: the tools every other course assumes you already know. Short, and worth doing early. |
11 | 10h 14m | 6h 49m |
| 13 | CS50's Understanding Technology — Harvard Hardware, the internet, multimedia, security and web development for people who use computers rather than program them. The shortest way in, and an easy one to recommend to someone else. |
6 | 5h 55m | 3h 57m |
Totals count available videos only and update as the playlists change. Private or deleted videos are excluded. Compare the top 5 side by side →
How to choose
- Never programmed? CS50x. If you'd rather go slower and in Java, Stanford's CS106A, then CS50x for the breadth.
- Want the ideas behind programming? SICP. It is old, in Lisp and unlike anything else here; that's the point.
- Heading for systems? The Missing Semester, then 6.004, CS 162 and 6.824, roughly in that order. CS 144 and CS 161 slot in anywhere after CS 162.
- Interested in AI? 6.034 for the ideas, CS50 AI for the code, then the free machine learning courses.
- Algorithms and data structures have their own list: MIT 6.006, 6.046J and the interview-prep courses.
Fitting a semester into a normal week
University lectures run from 50 minutes to two hours, and the problem sets take longer than the lectures do. Open a course here, give the planner your real weekly hours with the practice time included, and it will give you a date; tick lectures as you finish them and the remaining time updates. The guide to finishing a long course has the method, and watching at 1.5× works better for lectures than for anything you have to type along to.