Garry Tan (@garrytan) on X
My Thoughts
I use a very specific prompt to push Claude to check its work and do a lot of testing and thinking about perf and refactoring. I find I can do big features (4K LOC+ with full testing) in about an hour.
You might also like
share Why is Claude Code So Popular?
Why Claude Code is Gaining Popularity In this video, we explore why Claude Code has become so popular. Discussing its flexibility, simplicity, and robustness, the video touches on its foundation built on enduring, tried-and-true technologies like Unix commands and text files. The concept of the Lindy Effect is introduced to explain the longevity and future potential of these tools. We also compare Claude Code to other AI coding tools like VS Code and Cursor, highlighting its ability to efficiently serve both human programmers and AI agents. Additionally, we delve into the practical applications and intuitive design of Claude Code, making it accessible for everyone from beginners to experienced developers. 00:00 Introduction to Claude Code's Popularity 00:27 Boris Journey's Vision and Flexibility 01:50 The Lindy Effect and AI Coding Tools 02:48 Foundations of Claude Code 04:14 Comparing AI Coding Tools 05:25 The Future of Coding with Claude Code 07:34 Benefits of Unix Commands and File Systems 14:25 Practical Tips for Using Claude Code 16:47 Conclusion and Final Thoughts
share Chayenne Zhao (@GenAI_is_real) on X
I've been using Claude Code heavily lately, and while doing so, I've been casually watching the OpenClaw codebase evolve. What I've witnessed mirrors a pattern I've seen play out with every agent framework before it — and it's worth talking about. OpenClaw is a remarkable
share Andrej Karpathy (@karpathy) on X
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