AI Training & Quality Evaluation Specialist – Aether Project
Image
Bounding BoxWorked on the Aether project focused on improving AI model performance through data annotation, response evaluation, and quality review tasks. Responsibilities included analyzing AI-generated outputs for accuracy, relevance, instruction-following, reasoning quality, and safety compliance based on detailed project guidelines. Performed comparative evaluations between responses, identified factual inconsistencies, and provided structured feedback to improve training data quality for large language models. The project also involved handling multimodal and text-based tasks requiring strong attention to detail, critical thinking, and consistency across evaluations. Maintained adherence to strict quality standards, productivity targets, and annotation protocols while collaborating within a fast-paced AI training environment. Developed experience working with evolving guidelines, edge cases, and complex prompt evaluation workflows used in modern generative AI systems.