加 LINE 領取《自動化工作流研究筆記》
我們把企業導入 AI 自動化時最常卡住的十二個環節,整理成一份可以直接照著做的筆記。
AI & Edge Technologies
We help small and mid-sized companies actually use AI: an audit to find what is worth automating, a five-day PoC to prove it works, then a decision about scale. No shelfware, no all-or-nothing bets.
我們把企業導入 AI 自動化時最常卡住的十二個環節,整理成一份可以直接照著做的筆記。
Audit, prompts, proof of concept — validate first, scale later.
Look before you spend
A 60-minute remote audit of the workflows that eat the most time and people-power — and an honest answer on what AI can take over today, and what it cannot.
Make AI produce reliably, not by luck
Prompt sets and multi-step workflows designed for your actual tasks, stress-tested so they keep working when the data or the operator changes.
Five days to something that runs
No architecture decks — a working prototype built in five business days, validated against your real data.
Research and implementation, written down so the next attempt is faster.
Oct 23, 2026
The old site shipped every article with no title tag, description or canonical URL — and LaTeX that only appeared after JavaScript ran. Here is the move, and the three bugs that almost broke it.
Read article →
Aug 6, 2026
CALM compresses K discrete tokens into one continuous vector, cutting generation steps to 1/K. Notes on the autoencoder, the likelihood-free head, and what it actually buys you.
Read article →
Sep 16, 2025
How do you put physics into city-scale models? This survey proposes a taxonomy of three paradigms and seven methods, reviews eight urban subsystems, and names what is still missing.
Read article →Tell us about your workflow — we will tell you what can be automated today, and what cannot.