AI Evaluation, Prompt Engineering & Image Annotation Specialist
Performed high-precision image annotation and logical analysis for various machine learning and computer vision datasets. The project involved drawing precise bounding boxes and polygon outlines around specific objects, UI elements, buttons, icons, and textual components within catalogs of over three thousand images. Followed strict, multi-layered quality control guidelines to maintain an annotation accuracy rate of over ninety-nine percent, ensuring clean and structured datasets for model training. Additionally, evaluated logical consistency and prompt responsiveness for large language models, correcting output anomalies and refining guidelines to optimize model training performance.