AI Data Trainer & Linguistic Evaluator | Various AI and Data Service Platforms
Evaluated AI-generated Amharic content for grammar, fluency, and cultural appropriateness to improve overall dataset and model quality. Performed linguistic annotation and data labeling on large-scale language datasets to ensure correctness and consistency. Provided detailed error detection and correction feedback so labeling outcomes could directly enhance AI model performance. • Grammar and fluency evaluation for Amharic text • Linguistic annotation and data labeling on large-scale datasets • Error detection with corrective, feedback-oriented responses • Data collection & validation to support dataset quality