Who decides which large language model ranks first?
An article on large language model rankings, first published on the WeChat Official Account “杨攀同学” (Yang Pan Tongxue).
Writing
Thoughts on technology, products, agents, and developers.
Consensus and non-consensus in the first half of 2026
From context and model capabilities to software and organizational structures, based on a talk at the 43 College alumni gathering.
Software is frozen; the harness is alive
Starting with the Ship of Theseus, a discussion of harnesses that evolve at runtime and the relationship between software and agents.
Context is the agent: the next battle in AI products is over context
As models and harnesses converge, how does context define the limits of an agent’s capabilities?
The digital world: the web in the US, apps in China
Observations on differences between internet product formats in China and the US.
A tech entrepreneur’s year-end reading list for 2022
Reading notes and a year-end book list from 2022.
Taking Chinese cloud services global? Start with a status page
Cloud services going global and transparency about availability: why a status page matters.
A developer’s guide to asking questions in a community
How to ask effective questions and get help in technical communities.
No matching entries.