Instant, direct access to all 7 Stanford course playlists, textbooks, MuJoCo simulators, and open datasets.
Robot learning · 6 months · $0
Stanford charges $70,000 a year for robotics. The lectures are on YouTube for free.
Almost nobody watches them in the right order, so almost nobody builds anything. This is the order: math, then the robot's body, then how it sees, plans and learns, with one project every month.
119
free lectures
7
courses & seminar series
~95 h
of video, ~4 h a week
6
projects on your laptop
$0
no GPU, no robot needed
The path
- M0Python and your setupoptionalOnly if you've never coded. Skip it if you already write Python.~4 weeks, 1 h a day
- M1The math that everything else stands onVectors, matrices and least squares, taught as tools, not proofs.~18.5 h of video + practice
- M2The robot's bodyHow an arm moves, where its hand is, and how much force its motors need.~19 h of video + practice
- M3How a robot sees and sensesCameras, IMUs and lidar, and the neural networks that read them.~17 h of video + practice
- M4Planning and controlHow a robot decides what to do next, when it knows how the world works.~15.5 h of video + practice
- M5Teaching robots instead of programming themImitation learning and reinforcement learning, from Chelsea Finn and the AA203 team.~17 h of video + practice
- M6The frontier, and your first robotTalks from the people building robot AI, and one project that proves everything.~8.5 h of talks + final project
The problem isn't the content. It's the order.
A Stanford student doesn't pick lectures at random. There's a progression, and this path rebuilds it from free material only. No diploma, no grades, but the knowledge, and that's the part employers test in interviews.
- ×Start a famous ML course, feel smart for two weeks, never touch robotics again.
- ×Open a deep RL lecture, hear "Bellman" in minute four, close the tab.
- ×Watch a viral humanoid talk, then try to train a robot policy without knowing what a Jacobian is.
Taught by the people who build the robots
CS224R
Chelsea Finn
co-founder of Physical Intelligence
AA203
Marco Pavone
leads autonomous vehicle research at NVIDIA
CS223A
Oussama Khatib
trained generations of robotics engineers
guest lecture
Ashish Kumar
AI lead at Tesla Optimus
From 412 real job posts
The humanoid race isn't filtering by diplomas.
2 of 412
jobs make a PhD mandatory
160
jobs never mention a degree
$150K–$249K
median posted pay range
1 in 13
engineering posts mention ROS
Don't do it alone.
One checkpoint and one project in your inbox every week, so the path doesn't die in week two.