# Close the loop with verified improvement

Taiga Learning · Worksheet
https://taiga.training/en/lessons/continuous-improvement/

Use fictional or approved information. Do not put secrets in this worksheet.

## Learning objectives
- Connect an operational observation to a verifiable engineering change.
- Separate runtime recovery, workflow improvement, and model training.
- Measure a claimed improvement without weakening its evaluation.

## Exercise
Complete the feedback record in this lesson for the fictional memory leak. Define a baseline, acceptance test, regression checks, release boundary, production measurement, and owner. Add a rule for rejecting a faster but less correct export.

## Your response
- Scenario and scope:
- Assumptions and open questions:
- Proposed answer or decision, with reasons:

## Verify your response
| Claim or criterion | Evidence or test | Result or gap | Owner |
| --- | --- | --- | --- |
| | | | |
| | | | |
| | | | |

## Next action
- Action, owner, and date:
- When will you review this response?

## Principle to retain
The loop closes when a released change improves the observed outcome without violating other requirements. A proposed fix or merged PR is an intermediate result.

## Sources
- [Google SRE: Postmortem Culture](https://sre.google/workbook/postmortem-culture/)
- [Google SRE: Canarying Releases](https://sre.google/workbook/canarying-releases/)
- [Google SRE: Eliminating Toil](https://sre.google/workbook/eliminating-toil/)
- [Taiga docs: Maintaining](https://docs.tai.ga/operate/maintaining/)
- [Taiga docs: Initiatives](https://docs.tai.ga/deliver/initiatives/)

This worksheet supports learning. Completing it does not itself authorize a production change.