Built with Astro, hosted on AWS. Domain registered in 2024.
pinglin.tw is built with Astro, hosted on AWS. It measures traffic with Vercel Web Analytics.
| Technology | Version | First seen |
|---|---|---|
| JavaScript Frameworks & Libraries1 | ||
| 5.18.2 | 7 Nov 2026 | |
| Frontend Build Tools & CSS1 | ||
| – | 7 Nov 2026 | |
| CDN & Hosting2 | ||
| – | 7 Nov 2026 | |
| Analytics & Tag Management1 | ||
| – | 7 Nov 2026 | |
| Error Tracking & Performance Monitoring1 | ||
| – | 7 Nov 2026 | |
| Fonts & Icon Fonts1 | ||
| – | 7 Nov 2026 | |
| Translation & Localization1 | ||
| – | 7 Nov 2026 | |
Registration from the registry's RDAP record, the certificate the site serves, and its rank in public traffic lists.
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
Counted from every stack change we've recorded: of the sites we watched running one of this site's tools, how many went on to add another. A pattern across sites, not a forecast for this one.
From what the site says about itself, and account IDs it shares with other sites.
ns57.domaincontrol.comns58.domaincontrol.comgoogle-site-verification=Y7DROf7A9rNUS1Zcw9b96o6sNKJXQHxMtPOSWh3Gzow# All crawlers welcome, including AI/answer-engine bots # (Googlebot, Bingbot, GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot, ...). User-agent: * Allow: / Sitemap: https://pinglin.tw/sitemap.xml Sitemap: https://pinglin.tw/sitemap-index.xml
# pinglin.tw > Personal site of Ping-Lin Chang (張秉霖) — entrepreneur working on AI and data. Essays on AI, engineering, and building things, in English and Traditional Chinese. ## Pages - [About](https://pinglin.tw/about/): Who Ping-Lin Chang is - [Blog](https://pinglin.tw/blog/): All posts - [RSS feed](https://pinglin.tw/rss.xml) ## Blog posts - [The Patch You Do Not Wait For](https://pinglin.tw/blog/the-patch-you-do-not-wait-for/): Six security incidents in nine October days: PHP, GitLab, Node.js, LiteLLM, Railway and Zeabur. They share two moves: turn a published fix into an exploit, and steal a key that never expires. Here is how the agent-ops at A40 Labs answers: an exploited or Critical vulnerability gets a pull request within 4 hours and reaches production within 24. - [跟漏洞攻擊程式(exploit)賽跑的 patch](https://pinglin.tw/zh-tw/blog/the-patch-you-do-not-wait-for/): 十月中短短九天,PHP、GitLab、Node.js、LiteLLM、Railway、Zeabur 接連出事。仔細看,攻擊手法只有兩種:拿官方釋出的 patch 回推出漏洞攻擊程式(exploit),或是偷走一把永遠不會過期的金鑰。A40 Labs 的對策是:只要漏洞已遭利用或屬於 Critical 等級,agent-ops 在 4 小時內開出 pull request,24 小時內上 production。 - [Beyond OCR Scores: Where Document Parsers Fail](https://pinglin.tw/blog/beyond-ocr-scores/): A 1,250-page OCR benchmark exposes the scrambled reading order, missing tables, and broken formulas that RAG pipelines and AI agents inherit. I compare document parsers, OCR specialists, the open-weights Qwen 3.6, the hosted Claude Fable 5.1 and GPT-6 Astra, and Apple Vision to show where each fails—and what aggregate scores hide. - [OCR 分數之外:文件 parser 真正失敗的地方](https://pinglin.tw/zh-tw/blog/beyond-ocr-scores/): 用 1,250 頁沒有調校過的 held-out 頁面,把十二個 Reader 各自用作者自己的協定跑一遍:MinerU2.5-Pro + dots.mocr 的 composite、OCR 專用模型、open-weights VLM、兩個雲端 frontier 模型,以及每台 Mac 內建的免費 OCR。重點不是誰的平均分數漂亮,而是你的系統會在哪些頁面上直接失去資料。 - [The Cluster You Do Not Watch](https://pinglin.tw/blog/the-cluster-you-do-not-watch/): I almost never open Grafana any more. This is the operations architecture that made that true, on under US$10 a month of infrastructure — an agent that reads everything and writes nothing directly, alerts precise enough to act on, repairs that must prove they are safe before running, and one rule that turns out to be load-bearing: a check that cannot fail is not a check. None of it depends on which Kubernetes you run. - [The Shapes of Agent Memory – Files, Stores, and Experience](https://pinglin.tw/blog/the-shapes-of-agent-memory/): An agent that remembers across sessions can keep its memory as curated markdown files, as an auto-mined structured store, or as trained experience. I measured all of them: files against a structured store under one fixed model, a store-only head-to-head across the structured lineages, and an experience bank on the agentic benchmarks where the state of the art trains memory into the weights. - [Why Steering Works – The Theory Beneath X Engineering for AI Agents](https://pinglin.tw/blog/why-steering-works/): An agent run is stochastic gradient descent over solution space, and steering is how the true objective enters the loop. An optimization view and a Bayesian view that ground the industry ladder — prompt, context, harness, loop engineering — in foundation theory, and explain when a big model steering a small one beats distillation, and when it does not. - [Optimizing Transformer Model Serving Parameters – An Apple Silicon GPU Case Study](https://pinglin.tw/blog/optimizing-apple-silicon-gpu-for-transformer-inference/): One MacBook Pro M4 Max, an open-source MoE, and a hands-on exercise in optimizing it for inference: why the thing that collapses LLM throughput by 100× is prefill, not decode, and the two prefill-side fixes that got it back. - [優化 Transformer 模型服務參數 – Apple Silicon GPU 的案例](https://pinglin.tw/zh-tw/blog/optimizing-apple-silicon-gpu-for-transformer-inference/): 一台 MacBook Pro M4 Max、一個開源 MoE 模型,針對推論優化的一次實戰:為什麼讓 LLM 吞吐量崩跌 100 倍的是 prefill(而不是 decode),以及兩個 prefill 側的關鍵優化。 - [Collection Autofill at Scale](https://pinglin.tw/blog/collection-autofill-at-scale/): A design walkthrough of how Collection Autofill — a spreadsheet of AI-computed cells — stays fast, fair, and observable from a hundred rows to ten million. - [Collection Autofill 的規模化架構](https://pinglin.tw/zh-tw/blog/collection-autofill-at-scale/): 這篇文章談的是 Collection Autofill 的架構設計:一個把非結構化資料整理成 structured context 的產品介面,如何在幾百列到上千萬個 AI-computed cells 的規模下,仍然保持快速、公平且可觀測。 - [Reimagine Unstructured Data ETL](https://pinglin.tw/blog/reimagine-unstructured-data-etl/): Generative AI is revolutionizing software—how does it transform the way we utilize unstructured data like text, images, and videos? - [重新想像非結構化資料 ETL](https://pinglin.tw/zh-tw/blog/reimagine-unstructured-data-etl/): 生成式 AI 正在改變軟體,那它又會如何改變我們使用文件、圖片和影片等非結構化資料的方式呢?
| – |
| 7 Nov 2026 |