Data Annotator
Image
RLHFLanguage Data Contributor | AI Training Project | 2024 Took on self-directed annotation work to build practical experience in AI data pipelines. Tasks included labeling text for tone, intent, factual accuracy, and harmful content detection. Developed a sharp instinct for edge cases the ambiguous, context-dependent inputs that challenge even advanced models. Came away with a deeper appreciation for why human judgment remains irreplaceable in AI development. Built a consistent personal rubric for evaluating AI-generated responses across multiple quality dimensions including clarity, helpfulness, coherence, and potential for harm. Applied domain knowledge from a Biochemistry background and professional copywriting experience to assess outputs in both scientific and general language contexts. Engaged with RLHF concepts by ranking and rating model responses based on accuracy, tone appropriateness, and alignment with human values developing the kind of calibrated judgment that makes annotation work genuinely useful to model training teams.