mach-e

The first lecture of David MacKay's Cambridge course on information theory, pattern recognition and neural networks. It introduces the core ideas of his book Information Theory, Inference, and Learning Algorithms.

An excerpt of Richard Feynman's 1985 lecture Q&A about whether machines can think. He discusses the difficulty of defining what computers can never do and compares machine intelligence to machines that don't copy animals.

The final part of Dan Gelbart's 18-part Building Prototypes series, focused on mechanical design. It opens with a demonstration of stiffness and covers practical design principles for prototype machines.

John Gustafson presents posits, a proposed drop-in replacement for IEEE 754 floats that is simpler in hardware and more flexible in precision. The seminar includes a live demo and machine learning results.

Sacha Arnoud, Waymo's Director of Engineering, gives a lecture in MIT's 2018 deep learning for self-driving cars course on the rise of machine learning in autonomous vehicles. It covers how Waymo uses deep learning and how it handles tricky driving situations.

A recording of Richard Feynman's talk on the possibilities of manipulating matter at very small scales, widely seen as an early vision of nanotechnology. This is a later version of his 1959 "There's Plenty of Room at the Bottom" lecture.

Peter Norvig's UBC distinguished lecture on how massive amounts of simple data, such as for language modeling, can yield understanding that hand-built models could not.

Part 10 of Dan Gelbart's prototype-building lecture series, covering flexures: flexible joints that give precise, frictionless motion. It shows how a cheap piece of aluminum can become a nanometer-accurate mechanism.

Andrej Karpathy's first Neural Networks: Zero to Hero lecture, which builds micrograd, a tiny autograd engine, from scratch in Python. It explains backpropagation and neural network training step by step, assuming only basic Python and some high-school calculus.

Andrej Karpathy's three-hour-plus general-audience deep dive into how ChatGPT-style LLMs are built, covering the full training stack from pretraining to post-training. It also gives mental models for their quirks, like hallucination, and how to use them well.

Leo Dirac explains how transformer models have replaced LSTMs in NLP. He gives background from bag of words through RNNs and LSTMs, then a technical deep dive into multi-headed self-attention and positional encoding, with sample code.

Geoffrey Hinton's 2025 Royal Institution Discourse on what AI has learned from biological intelligence and how it might surpass it. He explains neural networks and argues that human language use resembles what backpropagation-trained networks do.

Sasha Rush surveys linear RNNs and state space models as replacements for attention in transformer-style models, in a talk for MLSys. It covers the foundational preliminaries and predates Mamba.

A university lecture from CS480/680 by Pascal Poupart on attention mechanisms and transformer networks, worked through on a chalkboard.

Bill Dally of NVIDIA gives a University of Washington lecture on trends in deep learning hardware. It covers numeric formats, specialized instructions, energy per operation and co-design of hardware and software.

Geoffrey Hinton gives an MIT talk on what is wrong with convolutional neural networks and introduces his capsules project. The description points to the capsule network papers.

The first lecture of Stanford's CS231n course on convolutional neural networks for visual recognition. It surveys the history and key challenges of computer vision, from the Cambrian explosion and Hubel and Wiesel to ImageNet.

David Shaw of D. E. Shaw Research presents Anton, a special-purpose machine that speeds up biomolecular simulations about a hundred-fold. The lecture covers molecular dynamics and how simulation helps understand how drugs bind.

A Bell Labs film in which Dr. J.N. Shive uses a mechanical wave machine to demonstrate reflection, standing waves, resonance and impedance matching. He includes some equations but relies mostly on real-world examples a layperson can follow.

The first lecture of Gilbert Strang's MIT 18.06 Linear Algebra course, covering the geometry of linear equations. It opens a full 35-lecture university course.

Noam Chomsky's 2011 University of Oslo lecture on Newton's contributions to the study of mind. He argues Newton abandoned mechanical explanation and lowered the bar for scientific intelligibility, with consequences for the limits of human understanding. Q&A starts at 45:33.

John Ousterhout's slides-and-voiceover lecture on Paxos, made for the Raft user study comparing how students learn Raft and Paxos. It is a full introduction to the Paxos consensus algorithm.

Gabriel Lebec's Fullstack Academy talk introduces lambda calculus syntax, the history of mathematical logic, and the essence of functional programming using JavaScript. It also covers combinators, currying and Church encodings.

The opening lecture of MIT 6.824 Distributed Systems (Spring 2020), introducing the course and core challenges of distributed systems. It begins a playlist that goes on to cover Raft and other consensus papers in Go.

The first lecture in Martin Kleppmann's 8-lecture distributed systems series from the University of Cambridge undergraduate course. It introduces the subject, and accompanying lecture notes are available.

Andrej Karpathy builds the GPT tokenizer from scratch, implementing byte pair encoding along with encode and decode. He shows how many odd LLM behaviors trace back to tokenization, and covers Unicode byte encodings along the way.

The opening lecture of Steve Brunton's Control Bootcamp, covering the kinds of control and why closed-loop feedback matters. It starts a series that follows the control chapter of the Brunton and Kutz data-driven science book.

Tom Wildenhain's guest lecture for CMU's Esoteric Programming Languages course shows that PowerPoint transitions and slides are Turing complete without macros, building up to a mind-bending fractal ending.

John Ousterhout lectures on the Raft consensus algorithm, made for the Raft user study comparing how students learn Raft and Paxos. It covers leader election and log repair.

Jeff Dean of Google describes his experiences building Google's systems and how its infrastructure evolved over time. The Stanford talk covers search, scaling tipping points and lessons from building large-scale software.

Ian Piumarta's 2007 Stanford EE380 lecture on building a dynamic programming environment from the ground up, from the Viewpoints Research Institute's work. It covers the design of a language and system with a recursive design, parsing through code generation, and the COLA architecture.

Anders Hejlsberg whiteboards how compilers are taught versus how modern ones are actually built. He explains the shift to compilers as services with incremental, error-tolerant, IDE-oriented design, as in Roslyn and TypeScript.

Mathematician Colin Wright demonstrates juggling and explains the mathematics of juggling patterns, including notation that led to new throwing patterns. It is a public presentation at the Museum of Mathematics aimed at a general audience.

Weber State's Professor Kelly explains the Toyota RAV4 Hybrid transaxle, including its planetary gearset eCVT and all nine operating modes. It also covers how the engine reaches nearly 40 percent thermal efficiency.

The first lecture of Stanford's Natural Language Processing with Deep Learning course, taught by Chris Manning and Richard Socher. It introduces NLP and word vectors, including SVD, skip-gram, CBOW, negative sampling and word2vec.

The first lecture of Michael Sipser's MIT 18.404J Theory of Computation course, covering course mechanics, finite automata, regular languages and regular expressions. It proves closure of regular languages under union and begins concatenation.

The first lecture of an MIT course on Douglas Hofstadter's Godel, Escher, Bach, covering themes like self-reference, recursion and incompleteness. The full lecture series is on YouTube.

Nathan Kutz presents the spatio-temporal dynamics of rotating detonation rocket engines, including combustion fronts that behave like solitons. He shows a bifurcation analysis of a multiscale model that captures the main features.

An MIT colloquium by Dennis Whyte on power exhaust and first-wall cooling in fusion reactors, including recent innovations and the SPARC and ARC designs.

Lecture 7A from MIT 6.001 (SICP) by Abelson and Sussman introduces the metacircular evaluator, a Scheme interpreter written in Scheme. It shows how evaluating a program is itself a program.