Focused on hardcore practical AI technologies (OpenClaw, ClaudeCode, LLMs, etc.) and HarmonyOS development. No hype—just real-world tips, pitfall chronicles, and productivity tools. Follow to transform workflows with code.
This article evaluates ByteDance’s Doubao-Seed-Evolving model, highlighting its weekly‑level updates, 1 M‑token context window, improved stability for long‑chain agent tasks, cost‑effective inference, and real‑world performance on a full‑project Android code‑refactoring case.
The article traces the evolution from Prompt Engineering to Context and Harness Engineering, introduces Loop Engineering as the next stage where AI runs autonomous cycles, explains its components, benefits, limitations, and offers guidance on when and how developers should adopt it.
The article evaluates Xiaomi’s open‑source MiMo Code, comparing its persistent memory, Compose mode, and /dream command against Claude Code, and provides a step‑by‑step hands‑on test that shows how MiMo Code’s engineering‑focused workflow can reduce rework despite a slower start.
Claude Code introduced dynamic workflows on May 28, 2026, enabling the AI to automatically decompose tasks, run dozens to hundreds of sub‑agents in parallel, and cross‑validate results, which acts like a project manager and can cut multi‑hour jobs down to minutes while offering guidance on when to use or avoid the feature.
This guide explains the concept of AI Agent Skills, walks through creating, installing, and managing a Skill—including file structure, YAML metadata, progressive loading, platform-specific considerations—and introduces a one‑stop development assistant that streamlines Skill development and deployment.
The article dissects the Agent Harness concept—covering its definition, three engineering layers, twelve production‑grade components, detailed orchestration loops, context‑management tricks, verification strategies, and how frameworks like Anthropic, OpenAI, LangChain, CrewAI and AutoGen implement these patterns, revealing why the same model can yield wildly different results.
Claude Code now supports a 1 million‑token context window, but effective use hinges on disciplined context management—choosing when to continue, rewind, clear, compact, or delegate to sub‑agents, and applying three core concepts of context windows, compaction, and context rot to avoid performance pitfalls.
The article explains that while prompt engineering helps AI answer single questions, building a robust execution environment—called Harness Engineering—is essential for agents to work continuously, reliably, and autonomously across complex tasks.
Anthropic’s Managed Agents transform the cumbersome agent runtime into a modular, production‑ready infrastructure by decoupling the brain, hands, and session layers, improving reliability, security, and performance while offering developers a clear path to build long‑running AI workflows.
The author explores Clawvard, an AI‑agent assessment platform that tests agents across eight dimensions, shares personal test results showing an initial A‑ rating with a critical retrieval weakness, details the customized improvement rules applied, and demonstrates a subsequent A+ rating, while also discussing the platform’s limits and practical use cases.
