Learn deeper, see farther.

Notes on AI, programming, and engineering—written through questions, experiments, and practice.

A LONG-RUNNING TECHNICAL NOTEBOOK · AI, CODE & ENGINEERING

A journey through mountains, sea, lake, and city.

What a journey it has been.

It began on the highlands of Gansu, where mountains fade into a vast loess horizon. From there, I travelled east to study by the East China Sea—years spent reading, writing, and learning how to ask better questions.

The road then turned south to West Lake. At the beginning of my working life, I had the good fortune to meet a mentor whose teaching and trust shaped the path ahead.

Now I live in Chengdu, in a quiet corner of the Land of Abundance. What I have learned and encountered along the way is kept here—for reflection, and for the road still ahead.

A traveller on the Loess Plateau looking towards a road winding east
Loess Plateau 01 / 04

Fields of inquiry

From model foundations to engineering practice, writing is organized around questions—not trends.

Recent writing

All writing

09. Agent Permission Is More Than Allow or Deny

Authorization decides whether one principal may perform one action on one resource under specific conditions

07. Are Parallel Tool Calls Actually Faster?

Concurrency shortens only independent waiting time and adds result joining, cancellation, and resource pressure

12. Context, Summaries, and Long-Term Memory Are Different

Longer retention does not create stronger knowledge; unverified memory carries old errors into new tasks

15. Evaluating an Agent Beyond Its Final Answer

A plausible answer does not prove that evidence is real, execution was authorized, or interruption recovery is correct

01. How System Prompts Are Assembled

The model does not receive one prompt. It receives context blocks assembled from different sources for different purposes.

06. How Tool Results Should Return to the Model

The model does not observe local execution; the runtime must carry evidence into the next turn