Azka Sheikh

I’m a detail-driven AI graduate with hands-on experience evaluating, labeling, and improving AI/ML model outputs across computer vision and NLP projects. I’m strong in written reasoning and in following quality rubrics, even under pressure.

In my AI Model Output Review internship, I reviewed and validated computer vision outputs (face detection), catching and correcting false detections. I also assessed and labeled object detection results with consistent criteria, and tested an NLP chatbot’s responses to reduce incorrect or irrelevant answers.

I bring careful judgment to prompt work as well—designing, iterating, and evaluating LLM workflows (certified, see below). I’m comfortable working asynchronously and unsupervised, because I’ve independently run a product business end-to-end and refined variations using performance data.

I’m seeking AI training, data annotation, and model evaluation roles where accuracy, consistency, and careful judgment matter most. I’m a fast learner of new guidelines and can deliver structured, rubric-based quality without needing heavy supervision.