# Evaluator-Optimizer

> Generator produces a candidate, evaluator scores it with feedback, generator revises — loop until criteria pass or budget exhausts.

- **Category**: Agentic AI
- **Subcategory**: Verification & Reflection
- **Canonical URL**: https://designpattern.fyi/patterns/evaluator_optimizer/

---

## Description
**Intent**: Climb predictably toward a quality bar by iterating generator and evaluator in a loop.

**Context**: Your generation task (code, translation, press release) has explicit criteria — tests pass/fail, rubric satisfied or not, glossary matches or doesn't. Single-shot generation plateaus below the quality bar you need, and you can afford a few model calls per output.

**Solution**: Generator produces a candidate. Evaluator scores it against criteria with actionable feedback. Generator revises using the feedback. Repeat until evaluator passes or max iterations reached.


## Use Cases
- Single-shot generation tops out below the quality the task requires.
- An evaluator can score candidates against criteria with actionable feedback.
- Iteration budget (max iterations or pass threshold) is acceptable in the latency model.





## Trade-offs


### Advantages

- Quality climbs predictably with iterations.

- Evaluator can be reused as an offline regression suite.




### Considerations & Drawbacks

- Cost = (generator + evaluator) × iterations — compounds fast.

- Convergence is not guaranteed; max-iterations budget is the safety valve.







---
**Reference**: [Original Source](https://www.agentpatternscatalog.org/patterns/evaluator-optimizer/)