# Neural Networks: Zero to Hero by Andrej Karpathy

29 Hacker News commenters recommended Neural Networks: Zero to Hero by Andrej Karpathy and 2 criticized it, November 2025 to November 2026. It was recommended in 9 of 13 months. It meets the Hacker Approved badge rule.

**Verdict:** Free video lecture series that builds neural networks from backpropagation up to a GPT-2 clone, praised for intuition. Caveat: some find it oversimplified.

## Why people recommend it

- Strong starting point for learners (6 comments): "all the skills you need to learn how to do this can be learned watching Andrej Karpathy’s zero to hero series" ([source](https://news.ycombinator.com/item?id=49423083)) "Supplement with the Karpathy 'Zero to Hero' video series if it suits your preferred learning style." ([source](https://news.ycombinator.com/item?id=48827960)) "I think Karpathys nn zero to hero is a good starting point." ([source](https://news.ycombinator.com/item?id=49087536))
- Builds intuition for core concepts (4 comments): "Does a great job of helping you build intuition for why things like gradient descent and normalization work" ([source](https://news.ycombinator.com/item?id=46490837)) "This series of videos was by far the best, most “intuition building”, highest signal-to-noise ratio" ([source](https://news.ycombinator.com/item?id=46486590)) "Karpathy has a unique talent for making complex topics feel approachable without dumbing them down." ([source](https://news.ycombinator.com/item?id=47203987))
- Approachable with little math (3 comments): "Anyone who watches the videos and follows along will indeed come up to speed on the basics of neural nets" ([source](https://news.ycombinator.com/item?id=46495380)) "lectures are as approachable as one could possible make them, no fancy math" ([source](https://news.ycombinator.com/item?id=46487072)) "I found the Karpathy videos very approachable." ([source](https://news.ycombinator.com/item?id=46486806))
- Covers backprop through GPT-2 clone (3 comments): "It starts with the fundamentals of how backpropagation works then advances to building a few simple models and ends with building a GPT-2 clone." ([source](https://news.ycombinator.com/item?id=46359861)) "You end up with a GPT-2.0 equivalent if you follow them to the end." ([source](https://news.ycombinator.com/item?id=50250351)) "found it very useful. Especially the back-propagation lectures." ([source](https://news.ycombinator.com/item?id=46488764))

## What people warn about

- Oversimplified, strips away depth (2 comments): "It's a useful series of videos no doubt, but his goal is to strip things down to basics" ([source](https://news.ycombinator.com/item?id=46498653)) "Karpathy has a great intuitive style, but sometimes it's too dumbed down" ([source](https://news.ycombinator.com/item?id=46487516))

## Alternatives hackers recommend

- [Math Academy](https://hackerapproved.com/p/math-academy.md): 35 recommend
- [Nand2Tetris](https://hackerapproved.com/p/nand2tetris.md): 31 recommend
- [CS193p SwiftUI course by Paul Hegarty](https://hackerapproved.com/p/cs193p-swiftui-course.md): 13 recommend
- [CS336: Language Modeling from Scratch](https://hackerapproved.com/p/cs336-language-modeling-from-scratch.md): 12 recommend
- [CS50 by Harvard](https://hackerapproved.com/p/cs50.md): 11 recommend

## Cite as

Hacker Approved, "Neural Networks: Zero to Hero by Andrej Karpathy", https://hackerapproved.com/p/neural-networks-zero-to-hero/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).