Search: twitter.com/karpathy | Heykuki News

Heykuki News

Top New Best Ask Show Jobs

Top New Best Ask Show Jobs

31.

Searching for the Agentic IDE (twitter.com)

33 points

3 months ago

32.

A few random notes from Claude coding quite a bit last few weeks (twitter.com)

32 points

4 months ago

33.

Llm.c Update (twitter.com)

31 points

2 years ago

34.

A lot of weird behaviors and problems of LLMs trace back to tokenization (twitter.com)

30 points

2 years ago

35.

Karpathy: "context engineering" over "prompt engineering" (twitter.com)

27 points

a year ago

36.

Andrej Karpathy: “Browsing the web, 2021” (twitter.com)

26 points

5 years ago

37.

Andrej Karpathy on X: "100% Software 2.0 computer.Just a single neural net (twitter.com)

25 points

2 years ago

38.

Andrej Karpathy on Hallucinations (twitter.com)

20 points

2 years ago

39.

Browsing the Web in 2021 (twitter.com)

18 points

5 years ago

40.

People have too inflated sense to "ask an AI" about something – Karpathy (twitter.com)

18 points

2 years ago

41.

Aspects to video generation that may not be fully appreciated (Andrej Karpathy) (twitter.com)

16 points

a year ago

42.

nanochat can now train GPT-2 grade LLM for –$73 (3 hours on single 8XH100 node) (twitter.com)

15 points

4 months ago

43.

Andrej Karpathy on X: RLHF is just barely RL (twitter.com)

15 points

2 years ago

44.

Andrej Karpathy on GitHub Copilot (twitter.com)

14 points

3 years ago

45.

Andrej Karpathy on X: LLM Knowledge Bases (twitter.com)

13 points

2 months ago

46.

The ongoing consolidation in AI is incredible (twitter.com)

13 points

4 years ago

47.

Andrej Karpathy: "LLM model size competition is intensifying backwards (twitter.com)

13 points

2 years ago

48.

Neural nets (*) across all areas are starting to look identical – a Transformer (twitter.com)

13 points

4 years ago

49.

Karpathy: We Should Bring Back RSS (twitter.com)

12 points

4 months ago

50.

Karpathy on Deepseek R1 (twitter.com)

12 points

a year ago

51.

First large diffusion-based LLM (twitter.com)

12 points

a year ago

52.

Karpathy Tweeted about OpenAI (twitter.com)

12 points

3 years ago

53.

Karpathy on Llama 3 (twitter.com)

12 points

2 years ago

54.

Andrej Karpathy reviews latest wide-release Tesla FSD version (twitter.com)

11 points

7 months ago

55.

Karpathy's comments on the Sutton/Dwarkesh podcast (twitter.com)

10 points

8 months ago

56.

Taking some time off to rest and travel after 5 years at Tesla (twitter.com)

9 points

4 years ago

57.

Andrej Karpathy: New art project. Train and inference GPT in 243 lines (twitter.com)

9 points

4 months ago

58.

Karpathy: Explaining llm.c in layman terms [tweet] (twitter.com)

9 points

2 years ago

59.

The real 30 under 30 imo you have never heard of (twitter.com)

8 points

2 years ago

60.

Andrej Karpathy: agentic AI coding has changed the world unrecognizably (twitter.com)

8 points

3 months ago