Open Internet by MindsNet

forrestchang/andrej-karpathy-skills

A tool providing guidelines to improve AI coding behavior, specifically for Claude Code, based on expert observations by Andrej Karpathy. It addresses common pitfalls like wrong assumptions and overcomplication in code generated by large language models. Best for: Developers and AI engineers working with large language models for coding tasks Use cases: Improving the accuracy and reliability of AI-generated code; Reducing the complexity of code produced by large language models; Enhancing the maintainability and readability of AI-generated codebases Highlights: Derived from expert observations by Andrej Karpathy on common LLM coding pitfalls; Provides four key principles for better AI coding: Think Before Coding, Simplicity First, Surgical Changes, and Goal-Driven Execution; Offers practical guidance for managing AI coding agents and improving code quality

Computing & Technology, Computer Science, Artificial Intelligence