University lectures & courseslecture
How to Speak
MIT OpenCourseWare
Documented HN sourceHN 21.0
Quick Take
HN treats it as the canonical, near-universally recommended lecture on giving talks and presentations.
Machine learning & deep learning
The spelled-out intro to neural networks and backpropagation: building micrograd
Andrej Karpathy · Discussed on HN from 2022 to 2025
Topics and ideas
- 01micrograd
- 02autograd
- 03backpropagation
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.
Read the notesSource context
Documented HN source
Makes backprop and neural nets simple and intuitive.
“I recommend watching Andrej Karpathy's recent videos on backprop and neural nets. Makes it really simple and intuitive.”Read the original HN comment
Geee · 2022-11 · 17 recommenders across 16 months
Keep exploring
Electronics & circuit designlecture
AT&T Archives: Similiarities of Wave Behavior (Bonus Edition)
AT&T Tech Channel
Documented HN sourceHN 19.0
Quick Take
HN repeatedly calls it the clearest introduction to waves and transmission-line behavior, better than modern explainers.
University lectures & courseslecture
Documented HN sourceHN 19.0
Quick Take
One of the most consistently recommended lecture series on HN, praised as fascinating, funny and eye-opening about human behavior across many scales.