I have hands-on experience in AI training, data labelling, prompt engineering, and model evaluation across multiple AI p
I have hands-on experience in AI training, data labelling, prompt engineering, and model evaluation across multiple AI platforms, including Alignerr, Outlier, Handshake, and Mercor. In these roles, I have worked on improving model quality by reviewing and scoring AI-generated responses against detailed rubrics, evaluating instruction-following, factual correctness, reasoning quality, and overall usefulness. I have created prompts and adversarial scenarios designed to uncover model weaknesses; contributed to supervised fine-tuning (SFT) and RLHF-style workflows; written grading guidance; and supported quality assurance processes to improve consistency and reliability of model outputs. My work has involved both technical and non-technical evaluation tasks, including comparing outputs across models and providing structured feedback that directly informed training improvements.