The Notebook · the week
The week of 9 November 2026
5 ideas · 9 November 2026 to 15 November 2026
-
Monday · 9 November 2026
llm streaming guardrails: the user read it before the filter
I am seeing a quiet gap in streamed chat answers. The model sends tokens to the screen as they arrive. The output guardrail runs on the finished answer. So the user reads every word before the check runs.
-
Tuesday · 10 November 2026
llm tokenization trailing space: one blank changed the answer
I am seeing a quiet bug in prompt templates. The template ends with a trailing space before the answer slot. Most tokenizers attach the leading space to the next word.
-
Wednesday · 11 November 2026
llm max tokens truncation: the json repair hid the cut
A model stops writing when it hits the max tokens limit. The API reports this in the finish reason, but many pipelines never read it. I saw one parse the cut text straight into JSON.
-
Thursday · 12 November 2026
llm eval statistical significance: the gain was inside the noise
A prompt change raised our eval score, and the team shipped it. A week later the gain had vanished on fresh questions. I am now wary of any single eval number.
-
Friday · 13 November 2026
ai agent memory: the old fact kept winning
I am seeing agent memory fail in a quiet way. An agent saves a fact about a customer. Months later the fact changes, and the agent saves the new one. Both versions now sit in the same store.