Engineering Decisions on 梦兽编程https://rexai.top/en/tags/engineering-decisions/Recent content in Engineering Decisions on 梦兽编程Hugo -- 0.163.3en梦兽编程Fri, 24 Jul 2026 23:50:00 +0800Claude Code Best Practices: The One Constraint That Governs the Workflowhttps://rexai.top/en/posts/2026-07-24-claude-code-best-practices/Fri, 24 Jul 2026 23:50:00 +0800https://rexai.top/en/posts/2026-07-24-claude-code-best-practices/A Meituan-style decision walkthrough of Anthropic's Claude Code Best Practices: context is scarce currency, verification gates, Plan mode, CLAUDE.md/hooks, session isolation, and scale all answer the same meta-question.llama.cpp, vLLM, SGLang: Three Inference Engines, Three Completely Different Pathshttps://rexai.top/en/ai/llm/inference-framework-comparison-2026/Wed, 22 Jul 2026 20:00:00 +0800https://rexai.top/en/ai/llm/inference-framework-comparison-2026/llama.cpp, vLLM, and SGLang aren't competitors in the same lane — they're built on entirely different design assumptions. Which one fits your workload? The answer depends on one metric you probably haven't measured.