Human-in-the-loop AI training support
Since December, I’ve worked with Outlier (a Scale AI platform) as a reviewer and audio tester, supporting human-in-the-loop AI training across both text and audio projects. In this role, I review AI-generated responses and annotator work for accuracy, clarity, reasoning quality, and adherence to guidelines, helping improve overall model performance. I’ve also worked with audio prompting and evaluation, reviewing speech-based interactions and edge cases used in training. My work draws on strong English language skills, including reading comprehension, writing, grammar, and clear communication, and spans a variety of writing styles and document types. Overall, the experience has strengthened my ability to apply consistent judgment, follow detailed standards, and provide practical feedback in a fast-changing AI training environment.