Jiayuan (JY) Zhang on X: "I let Claude Code turn @karpathy's post into agent skills. It first generated a bunch of skill files and around 800 lines of descriptions. Then I let it use these agent skills to review itself. Boom, it cut itself down to 70 lines of clean, solid instructions. https://t.co/
Quick Answer
Jiayuan Zhang utilized Claude Code to transform Andrej Karpathy's post into agent skills, resulting in an initial output of 800 lines that was refined to 70 lines of concise instructions.
Quick Take
Jiayuan Zhang utilized Claude Code to transform Andrej Karpathy's post into agent skills, resulting in an initial output of 800 lines that was refined to 70 lines of concise instructions. This shift highlights the growing efficiency of LLMs in coding workflows, moving from 80% manual coding to 80% agent-driven coding.
Key Points
- Claude Code generated 800 lines of agent skill descriptions from a single post.
- The output was refined to just 70 lines of clear instructions.
- Zhang's workflow shifted to 80% agent coding, enhancing productivity.
- This reflects the increasing capabilities of LLMs in software development.
- The transformation showcases the potential of AI in automating coding tasks.
📖 Reader Mode
~1 min readPost
I let Claude Code turn
@karpathy's post into agent skills. It first generated a bunch of skill files and around 800 lines of descriptions. Then I let it use these agent skills to review itself. Boom, it cut itself down to 70 lines of clean, solid instructions. github.com/forrestchang/a…
A few random notes from claude coding quite a bit last few weeks. Coding workflow. Given the latest lift in LLM coding capability, like many others I rapidly went from about 80% manual+autocomplete coding and 20% agents in November to 80% agent coding and 20% edits+touchups in
Don't miss what's happening
People on X are the first to know.
— Originally published at x.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from WebSearch (Tavily)
See more →Stop just chatting with AI. Learn to build production-ready software in ...
The 2026 Bootcamp offers hands-on training in building production-ready software using Generative AI, LLM applications, and AI agents, emphasizing practical skills over casual interaction with AI. Participants will learn to develop applications like Cursor AI, preparing them for real-world challenges in AI development.