AI training
I have one full year of practical experience in AI training and data labeling through Prolific, where I consistently completed paid studies involving multimodal evaluation and annotation. This included reviewing and annotating images (e.g., interpreting content, applying labels/rules, spotting inconsistencies for computer vision training), audio (e.g., evaluating speech, generating/annotating sound clips for speech recognition or quality checks), and videos (e.g., frame-by-frame tracking, labeling actions/events, providing feedback on sequences for video understanding models)—all contributing to reinforcement learning from human feedback (RLHF), model evaluation, ground truth creation, and AI safety/validation. These tasks demanded precise guideline adherence, attention to detail across media types, clear reasoning in feedback, and consistent quality judgments under varying complexity, building transferable skills in human-in-the-loop data refinement for text, visual, and auditory AI systems.