You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
UX improvements and fix broken markdown tables (#9)
UX changes:
- ToC right rail: active heading now has a left-border indicator
- Keyboard nav: ArrowLeft/ArrowRight navigate between chapters
- Reading progress bar: 2px height with stronger contrast
- Search: ArrowUp/ArrowDown/Enter keyboard navigation in results
- Mobile sidebar: overlay fades in/out in sync with slide transition
Content fixes:
- Fix markdown tables with rows wrapping across multiple lines in
02-capability-jump, 27-enterprise-adoption, 32-backpressure,
33-adoption-playbook, and appendix-b-glossary
Co-authored-by: Ona <no-reply@ona.com>
Copy file name to clipboardExpand all lines: src/content/chapters/32-backpressure.mdx
+3-6Lines changed: 3 additions & 6 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -180,12 +180,9 @@ Backpressure isn’t binary. It’s a dial.
180
180
181
181
| Setting | Symptom | Fix |
182
182
| --- | --- | --- |
183
-
| **Too little** | Hallucinations pass through. Agent output looks clean but is subtly
184
-
wrong. Human catches logic errors. | Add more automated checks. Strengthen test coverage. |
185
-
| **Too much** | Feedback loop is too slow. Agent waits 20 min for tests. Speed
186
-
advantage lost. | Parallelize checks. Run only affected tests. Cache build artifacts. |
187
-
| **Just right** | Agent self-corrects in 1-3 iterations. Human reviews architecture and
188
-
judgment calls only. | Target: full cycle in < 2 minutes. |
183
+
|**Too little**| Hallucinations pass through. Agent output looks clean but is subtly wrong. Human catches logic errors. | Add more automated checks. Strengthen test coverage. |
184
+
|**Too much**| Feedback loop is too slow. Agent waits 20 min for tests. Speed advantage lost. | Parallelize checks. Run only affected tests. Cache build artifacts. |
185
+
|**Just right**| Agent self-corrects in 1-3 iterations. Human reviews architecture and judgment calls only. | Target: full cycle in < 2 minutes. |
189
186
190
187
## Measuring backpressure effectiveness
191
188
How do you know if your backpressure is working? Track three metrics.
0 commit comments