# Best practices

## Evaluations

* Pick 3-5 benchmarks that match your task; resist "more"
* Re-evaluate quarterly as the model frontier moves
* Use spaces for stable configs; one-offs for exploration

## Judges

* GEPA-optimize against ≥30 labels
* Hold out a validation set
* Watch for over-fit (validation agreement << training)

## Continuous evaluation

* Sample production traffic; full corpus is rarely worth it
* Anchor with deterministic rules; reserve judges for residual subjective bar

## Govern

* Wire CI gates after you've trusted the eval interactively
* Per-feature tolerances — not every feature has the same bar

## See also

* [Workflow](/1.-introduction/the-stratix-workflow.md)
* [Concepts](/13.3-concepts-library-+-architecture/concepts-library.md)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.layerlens.ai/12.-support/best-practices.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.