Yes, I do
Yes, I do. I have practical experience working on data annotation and AI training projects, specifically focusing on text classification, Named Entity Recognition (NER), and Reinforcement Learning from Human Feedback (RLHF). In my experience, I have reviewed, tagged, and cleaned datasets to train models to better understand human intent and context. This included grading AI-generated responses for factual accuracy, identifying harmful biases, and evaluating overall helpfulness. Because of my background in computer engineering, I don’t just look at it as a repetitive task—I look at it through the lens of data integrity. I understand that the model's output is only as good as the 'ground truth' data we feed it, so I pride myself on maintaining a 98% or higher accuracy rate, even when handling complex or ambiguous labeling guidelines.