AI Data Quality Reviewer & STEM Trainer
Text
As an AI Data Quality Reviewer, I evaluated and annotated thousands of AI-generated responses across STEM domains. I applied detailed rubrics to assess model outputs and authored high-quality prompt-response pairs used for fine-tuning LLMs. I also conducted red-teaming evaluation for model safety, bias, and error testing.
• Rated scientific and technical outputs for accuracy, coherence, and relevance.
• Used RLHF pipelines to directly improve large language model performance.
• Maintained >94% inter-annotator agreement rate on concurrent annotation projects.
• Refined annotation guidelines and contributed structured feedback to iterative product improvement.
As an AI Data Quality Reviewer, I evaluated and annotated thousands of AI-generated responses across STEM domains. I applied detailed rubrics to assess model outputs and authored high-quality prompt-response pairs used for fine-tuning LLMs. I also conducted red-teaming evaluation for model safety, bias, and error testing. • Rated scientific and technical outputs for accuracy, coherence, and relevance. • Used RLHF pipelines to directly improve large language model performance. • Maintained >94% inter-annotator agreement rate on concurrent annotation projects. • Refined annotation guidelines and contributed structured feedback to iterative product improvement.
2023 - Present