Back
Cost:L3
|
Reward:L+6
gpt-4.1-mini

Translation Quality Pipeline

hardMulti-Agent8 tests

The localization team is overwhelmed. You're building an AI translation pipeline: one agent does the initial translation, another reviews and polishes it. The final output should be just the clean translated text — no notes, no "here's my translation," just the words.

Agent Prompts

Translator

Translates the input text to the target language.

Reviewer

Reviews the translation and outputs the final polished version.