AI Data Annotator & Evaluator
I evaluated and ranked AI-generated responses in both Amharic and English, focusing on factual accuracy and logical consistency. The tasks involved pairwise comparison of both text and image outputs, identifying errors, and providing structured feedback. I adhered strictly to complex annotation guidelines and maintained high levels of quality assurance. • Evaluated AI-generated text and images for accuracy • Ranked model outputs using pairwise comparisons • Identified instruction violations and inconsistencies • Provided detailed written justifications for decisions