All Articles

20 articles

杨正武 April 13, 2026

Hermes vs OpenClaw: The Memory Problem Nobody Wants to Solve

While everyone debates whether Hermes Agent or OpenClaw is better, a more fundamental question keeps getting dodged: how should Agent Memory and Context actually be designed?

#AI Agent #Memory #Context Design #Agent Architecture

李雪涛 April 8, 2026

How Do You Actually Manage Changes in the AI Era? (Opening the Discussion)

Traditional change management relies on friction to control change frequency. AI eliminated that friction. Change management has shifted from a human process problem to a context engineering problem. Sharing our practice and unsolved challenges in structured anchoring within an SDD system.

#ai-coding #spec-coding #change-management

李雪涛 April 8, 2026

What Should Project Management Look Like in the AI Era

AI drove code production costs to the floor, but project management costs actually went up. Traditional PM manages whether you finish on time; AI-era PM manages whether you're building the right thing. These require fundamentally different tools.

#ai-coding #project-management #team-management

李雪涛 April 8, 2026

Is Harness What Comes After SDD?

From prompt engineering to Harness Engineering, AI programming has been doing one thing: humans move further from code while gaining stronger control. OpenAI's 3 engineers guiding Codex to 1M lines of code reveals Harness's decisive role in AI programming reliability.

#ai-coding #harness #sdd #agent

瑞典马工 April 6, 2026

Stop Learning AI. Start Using It.

The best way to learn AI is to skip the learning and solve a real business problem. A $40/month store assistant case shows why starting from problems beats starting from technology.

#ai-practice #north-star-metric #business-first

杨正武 April 2, 2026

Feature Workflow v3: From Templates to Tailored Design — AI Workflows Are No Silver Bullet

Feature Workflow v3 upgrades to a Command + Agent + Skill three-layer architecture. The core insight: template-based workflows are not silver bullets — every project needs purpose-built Skill design.

#ai-agent #skill-design #workflow #ailock-step #customization

付权智 March 27, 2026

System Infrastructure for the Agent Era: Recap of the First AgenticOS Workshop

A summary of the first AgenticOS Workshop at ASPLOS 2026, exploring agents' dual role: as new users demanding fault tolerance, exploratory execution, and resource control; and as new builders enabling specialized software revival and LLM-assisted system management.

#systems #agent-infrastructure #agenticos #workshop #os

杨正武 March 23, 2026

Enterprise Software in the Agent-First Era: From SOP to Skill

Exploring the core of Agent-First transformation: Skill codification, security design, and commercialization methodology, drawing from cross-border e-commerce practices.

#agent-first #skill-design #enterprise-transformation #security

杨正武 March 13, 2026

AILock-Step Protocol: An Engineering Solution to AI Agent Hallucination Skipping

A strict linear execution protocol based on STP state anchors. Eliminates hallucination skipping via physical jumps, enabling checkpoint-resumable AI Agent development with absolute idempotency.

#AI Agent #工程实践 #执行协议 #MCP

李雪涛 March 9, 2026

AI-Native Org Transformation: Six Key Insights

AI transformation isn't patching old systems with AI — it's rebuilding organizational DNA. The core is human-AI symbiosis, small-team warfare, value chain compression, platform capabilities, and new management models. Miss any one and you're dead.

#ai-native #org-transformation #enterprise-management

付权智 March 7, 2026

Agent Management: The Deterministic Boundaries of AI Coding

AI Coding is not a technical problem — it's a management problem. From requirements to execution to acceptance, building explicit contracts at each stage is the deterministic boundary of AI Coding. A milestone summary from the Agent Management forum.

#ai-coding #agent-management #methodology #software-engineering

瑞典马工 February 15, 2026

How to Ensure Quality in AI Software Engineering

AgentsZone's roundtable of 19 AI coding pioneers explores how probabilistic LLM systems can ensure code quality through deterministic constraints and independent validation.

#quality-control #methodology

付权智 February 14, 2026

End-to-End Evolution for Unknown Failures: CloudMate's Adaptive Path Exploration

CloudMate handles tens of thousands of failure analysis requests per week. This article dissects its evaluate-mutate-backtest loop — how to build a self-evolving Agent system that adapts to unknown failures in a constantly changing production environment.

#aiops #cloudmate #agent #operations

瑞典马工 February 8, 2026

The Missing Role in Software Engineering: Why We Need a Code Supervisor

Lovable, valued at billions, had user debt amounts, home addresses, and API keys extracted in 47 minutes. Software has product managers, architects, developers, testers, and ops — but lacks one role that civil engineering calls a "supervisor."

#quality-control #team-structure

瑞典马工 January 26, 2026

Learning from PingCAP: Building Your Company's AI-Native Software Team

A response to PingCAP CTO Edward Huang's "Vibe Engineering 2026.1." Huang rewrote TiDB's PostgreSQL compatibility layer with AI, achieving near-production quality code. But he spends 90% of his time evaluating AI output. How do ordinary teams replicate this?

#methodology #team-structure

付权智 January 3, 2026

How AIOps Systems Should Evolve with Software: Lessons from CloudMate

Knowledge-based AIOps systems face a fundamental challenge — software keeps iterating and knowledge expires. CloudMate made a systematic exploration on two fronts: how to keep a knowledge base current with fast-moving code, and how to ensure software evolution doesn't break AI ops effectiveness.

#aiops #cloudmate #knowledge-base #operations

付权智 December 19, 2025

Will LLMs Disrupt Traditional Operations? A Survey of AI-Driven Ops

Surveying recent AI operations research to answer three questions: what has AI ops already achieved, where are the current limits, and what comes next. From Microsoft's RCACopilot to NeurIPS's Stratus, a deep dive into Context Engineering and LLM Agent opportunities and challenges.

#aiops #llm #research #operations

瑞典马工 December 15, 2025

Forward Deployed Engineer: Silicon Valley's Rebranded Delivery Consultant

In 2025, Silicon Valley is swept up in a Forward Deployed Engineer (FDE) hiring frenzy. OpenAI, Anthropic, Databricks and others are all recruiting FDEs. Is this genuinely new, or just a fancy name for the traditional delivery engineer?

#ai-thinking #industry-analysis

瑞典马工 August 9, 2025

Alibaba's qwen-code Was Too Hasty

Alibaba's qwen-code CLI is a fork of Google's Gemini CLI. Legally fine under Apache 2.0, but from a product management perspective, this was a rash decision. Users discovered the /init command still generates GEMINI.md instead of QWEN.md — a leftover from an incomplete fork audit.

#ai-coding #industry-analysis

瑞典马工 July 26, 2025

Using Iterative Methodology to Deliver Industrial-Quality Software with AI

Complaints that AI-written code is unusable? The problem is not correctly applying iterative methodology. Iteration means repeatedly approximating a fixed goal through multiple attempts. In software development, iteration applies to every stage — from requirements to implementation and testing.

#methodology #ai-coding