Andrej Karpathy: the videos Hacker News recommends
Not the channel's most viewed videos: the ones HN commenters kept linking with real commentary. The top pick, "The spelled-out intro to neural networks and backpropagation: building micrograd", hN treats it as the best starting point for really understanding backprop and neural nets, because Karpathy explains the why with intuition and builds everything in code.
- lecture recommended in 16 different months by 17 HN users, 2022–2025HN treats it as the best starting point for really understanding backprop and neural nets, because Karpathy explains the why with intuition and builds everything in code.
- lecture recommended in 8 different months by 10 HN users, 2025–2026HN's most frequent recommendation for a clear, approachable foundation in how LLMs work without heavy math.
- tutorial recommended in 8 different months by 14 HN users, 2023–2025It is the go-to recommendation for learning how transformers and LLMs actually work, praised for its clarity and Karpathy's teaching.
- talk recommended in 6 different months by 9 HN users, 2023–2026HN recommends it as the dense, approachable starting point for understanding LLMs and often cites it for specific claims about scaling and training.
- lecture recommended in 5 different months by 5 HN users, 2024–2025Recommended as the best deep dive on tokenizers, worth the two hours for anyone working with LLMs.
- tutorial recommended in 2 different months by 2 HN users, 2024HN recommends it as a long but worthwhile way to demystify how language models work technically.
- course recommended in 1 different month by 2 HN users, 2016–2023HN lists it as a free, brilliant introduction to deep learning and computer vision, often within ML self-study plans.
Questions
Which Andrej Karpathy video does Hacker News recommend most?
The spelled-out intro to neural networks and backpropagation: building micrograd: recommended in 16 different months by 17 HN users between 2022 and 2025.
Is this every Andrej Karpathy video?
No. Only videos that at least three different HN users linked, in at least three different months, with at least a sentence of their own commentary. That filter is what makes the list useful.