AI Briefing · 09-11New
Mingyu's Library — a personal libraryUpdated September 11, 2026
One AI topic,
explained every day
A daily AI briefing, a daily deep dive and a Claude Code course — three threads updated every day, plus long-form notes and tools I build.
- active days (last 30)
- 21 active days (last 30)
- pieces
- 254 pieces
- long-form notes
- 70 long-form notes
Three threads · refreshed daily
Today
AI Deep Dives · 09-11New
托管 Agent 运行时:OpenAI Agents API 深度解析 | Managed Agent Runtime: A Deep Dive into the OpenAI Agents API
ReadClaude Code Course · 08-05
D8 · Hooks
For the first seven lessons, every way you influenced Claude was "saying it to the model": CLAUDE.md, plans, Skills. Hooks take a different road — no…
ReadHand-picked, start here
Signature picks
AI
Multiple Sessions or Multi-Agent? I ran 27 agents to answer that
An orchestration study with real experiments: 6 probes to pin down where a subagent's isolation boundary sits, and 13 agents building the same game three different ways, each build verified in a headless browser. The counterintuitive result — multi-agent buys width inside one task, multiple sessions buy length across time and total context; they're orthogonal axes. What actually decides whether parallel development works is an interface contract everyone follows (without one, 2 of 5 assemblies went straight to a blank screen).
2026年7月28日Read
Writing
iOS App 第一次出海:归因与投放接入入门指南
整理自 Adjust 线上课堂:iOS 归因为什么难、ATT 与 SKAN 两套独立框架怎么理解和配置、Google(ICM)/ Meta(MAE/AEM)/ TikTok(SAN)三大渠道接入路径、归因窗口对齐建议与该看的指标,收尾附首次出海接入七步清单
2026年7月8日Read
AI
112 Days Late: I Built a Product Anthropic Had Already Shipped
A field report with the full data. I spent 22 days and 215 commits building a macOS app to manage Claude Code sessions — Anthropic had shipped most of it 112 days before my first line, and my one competitive check scoped itself around the mobile line and missed the desktop redesign entirely. Includes the 48-feature audit (18 replaced / 15 overlapping / 15 unique), the review economics (~190 agents for ~170 findings, 33% hit rate at one stage), six dated incidents including a symlink that destroyed an 823MB corpus, and a full data appendix with timestamps.
2026年8月26日Read
AI
An afternoon on Supabase: pgvector, a 44x query plan, and two RLS lessons
Supabase 实战英文复盘:pgvector 44x 查询优化与 RLS 双闸门
2026年7月27日Read
AI
Agent Runtime Architecture: The Reasoning Loop (Vector)
English vector diagram of the Agent runtime architecture — Runtime harness, Think/Act/Observe loop, on-demand resources, tool-result feedback
2026年7月1日Read
Published every day
Daily columns
AI Briefing
daily55 itemsToday's AI world in five minutes: releases, agent engineering, research and industry.
Latest: AI 每日简报 · 2026年9月11日
AI Deep Dives
daily51 itemsOne technical topic a day, written up as a thorough deep-dive document.
Latest: 托管 Agent 运行时:OpenAI Agents API 深度解析 | Managed Agent Runtime: A Deep Dive into the OpenAI Agents API
Claude Code Course
daily78 itemsA 14-day systematic course plus practical tips, updated daily from the official docs.
Latest: D8 · Hooks
Built, not just written
Apps & tools
App StoreSuiRead (iOS app)
An HTML reader I built. Download any article here, import it, and read offline with highlights and saved progress.
iOS阅读器App Store
Open
WeChat Markdown formatter
Turn Markdown into paste-ready WeChat article styling — live preview, multiple themes.
Markdown公众号排版
Open
Claude Code learning hub
Both a daily column and a build: a static learning hub with a 14-day course and searchable tips, updated daily.
Claude Code教程每日更新
About Mingyu
Who writes this?
I'm Mingyu — 10+ years in mobile tech. This site is my AI learning base: automated daily pipelines plus long-form notes and tools I build by hand.
About me