Back to index
OtherSelf-paced. Hundreds of videos grouped into playlists; the data-driven control playlist alone is about 46 videos, and the physics informed machine learning series runs five core parts plus follow-ups on PINNs, neural ODEs, neural operators and symbolic regression·Free on YouTube. The companion book, Data-Driven Science and Engineering (Brunton and Kutz, Cambridge University Press, 2nd edition 2022), is paid; the MATLAB and Python code and data are free at databookuw.com

Steve Brunton Lectures (Eigensteve: Data-Driven Science, Physics Informed ML and Control)

4.6

The best free material on machine learning for engineers and physical scientists. Brunton explains hard linear algebra and dynamics more clearly than almost anyone, though this is not a route into LLM or product AI work.

What We Liked

  • Exceptionally clear teaching, delivered with his signature lightboard style that keeps your eyes on the maths
  • Covers ground most ML courses ignore entirely: dynamical systems, control, reduced order models and physics constrained learning
  • Tied to a proper textbook with free MATLAB and Python code, so you can go from video to working notebook
  • The physics informed ML series is a balanced, sceptical overview of PINNs and neural operators rather than hype
  • Totally free, and individual videos stand alone well, so you can dip in for one topic

What Could Be Better

  • Almost nothing on LLMs, agents or the generative AI topics most people searching for AI courses want
  • No exercises, grading or certificate on YouTube itself; structure is up to you
  • Assumes comfort with linear algebra and differential equations, so true beginners will struggle by the second playlist
  • Older playlists lean on MATLAB, which is a nuisance if you only work in Python
  • The book costs money and parts of the series only make full sense alongside it

Detailed review

Steve Brunton is a mechanical engineering professor at the University of Washington, and his YouTube channel (Eigensteve) has become the default reference for a corner of machine learning that the big MOOCs barely touch. The core of it follows his textbook with Nathan Kutz, Data-Driven Science and Engineering, and moves from the singular value decomposition through Fourier and wavelet methods, sparse regression, dynamic mode decomposition and SINDy, into control theory and reinforcement learning. More recently he has added a physics informed machine learning series that walks through the stages of building a model (choosing what to model, curating data, designing an architecture, crafting a loss function) before covering PINNs, Fourier neural operators, DeepONets and Hamiltonian networks. What makes it work is the teaching.

He writes on a lightboard facing the camera, talks at a steady pace and keeps returning to why a method exists rather than just how to run it. I have watched his SVD videos several times and they are still the clearest explanation of the topic I know, better than most university lectures I sat through. I also like that the physics informed series is honest about where PINNs fall short, which is rare in a field full of breathless papers. The limits are worth being clear about.

This is applied maths for scientists and engineers, not an AI career course. You will not learn anything about prompting, RAG or deploying LLM apps here, and there is no certificate to put on LinkedIn. The playlists are not sequenced as a single course either, so you need some self-discipline, ideally with the book open beside you. Some of the older material assumes MATLAB, although the Python port of the book code helps.

For the right person, though, nothing paid comes close. If your work involves fluids, structures, robotics, energy systems or any physical process, this is the free series I would recommend first, and the book is one of the few ML textbooks I would actually pay for.

[ final ]

The verdict.

If you are an engineer, physicist or applied maths student who wants to use machine learning on real physical systems, start here before paying for anything. If you want to build chatbots or get an AI engineering job, look elsewhere.