Posts
All the articles I've posted.
-
MiroFish: A Swarm Intelligence Prediction Engine a College Senior Built in 10 Days with Vibe Coding
A deep dive into the architecture and design of MiroFish, a swarm intelligence prediction engine. Built by a college senior in 10 days with Vibe Coding, it topped GitHub Trending and landed a 30M RMB investment from Shanda. Here's what it got right, technically.
-
When AI Agents Learn to Pay: A Deep Dive into the x402 Protocol and Agent Payment Infrastructure
How Coinbase's x402 protocol revived the HTTP 402 status code to let AI agents pay for API calls with stablecoins. From protocol design to hands-on code, from the Stripe/Mastercard competitive landscape to the reality of $28K daily volume — a full breakdown of the agent payments space.
-
Browser-Native AI in Practice: The Complete Guide to Chrome Built-in AI APIs
A deep dive into Chrome's built-in Gemini Nano model and the browser-native AI APIs (Prompt, Translation, Summarization) — from technical architecture to production practice, showing how to build privacy-first local AI apps, with complete code examples and performance analysis.
-
OpenClaw Automation in Practice: Building a 24/7 AI Assistant with Cron + Heartbeat
A deep dive into OpenClaw's Cron Job and Heartbeat mechanisms—from choosing between them to engineering practice—with a real production case study covering error handling, state management, and cost optimization.
-
The 2026 AI Model Landscape: Hands-On Comparison of 12 Leading Models from China and Abroad
Hands-on benchmarks of 12 AI models (GPT-4o, Claude 3.5, Gemini 2.0, Qwen 2.5, GLM-4, Kimi, and more) across 6 real-world scenarios including code generation, Chinese writing, and reasoning. Includes performance scores, monthly cost comparisons, and a decision tree to help you pick the right model.
-
Building AI Agents with Long-Term Memory: From Design Patterns to Production
A deep dive into building episodic, semantic, and procedural long-term memory for AI Agents, with a complete technical architecture, code implementation, and production optimization strategies.
-
AI Agent-Driven Development: The Paradigm Shift from Tools to Workflows
How AI Agents are evolving from "assistive tools" into "collaborative partners" and reshaping the modern developer workflow. Based on real project experience, a deep dive into three core capabilities — context awareness, proactive execution, and tool orchestration — with complete code examples.
-
The AI Agent Skills Standardization War: Architecture, Security, and Ecosystem Evolution
A deep dive into the technical architecture, security models, and governance mechanisms of MCP, Agent Skills, and Skills.sh — unpacking the design-philosophy clash behind real-world security incidents, with a practical selection guide for enterprises and developers.