📄 中文摘要
在日常开发中，使用 ChatGPT、Claude 和 Cursor 等大型语言模型（LLMs）已成为许多开发者的常态。这些工具在短期任务中表现良好，如生成函数、调试和解答问题。然而，作者希望探索更长期的项目，超越一次性脚本的限制，构建一个持续的个人知识管理系统。该系统旨在帮助跟踪专业知识在不同领域之间的联系，包括研究主题、技能发展、决策过程及项目经验，以避免散乱笔记和遗忘的教训。
尝试用大型语言模型构建真实项目时会发生什么
出处: What Happens When You Try to Build Something Real with LLMs
发布: 2026年2月16日
📄 English Summary
What Happens When You Try to Build Something Real with LLMs
Many developers rely on large language models (LLMs) like ChatGPT, Claude, and Cursor for daily tasks such as function generation, debugging, and problem-solving. These tools excel in short, well-defined tasks. However, the author sought to explore a different approach by embarking on a long-term project rather than a one-off script. The project involved creating a personal knowledge management system designed to track the connections between professional knowledge across various domains. This system aims to document research threads, skill development, decision-making processes, and lessons learned from projects, thereby preventing valuable insights from being lost in scattered notes.