At SpaceXAI, I progressed to L1/L2 Quality Reviewer responsibilities across three AI projects within my first month. I reviewed over 100 contributor submissions daily and maintained an average internal quality score of 6.9/7.
I evaluate LLM responses through multi-turn conversations, side-by-side comparisons, and written preference justifications. I also assess speech AI transcription, voice naturalness, pronunciation, and linguistic accuracy.
For Grok, Starlink, Tesla, X, and SpaceXAI, I performed English-to-Italian localization and linguistic QA. At Outlier, I design prompts and conversations, compare model outputs against rubrics, and review contributor submissions.
My background includes Italian linguistics and dialectology, with a focus on Sardinian and Friulian, as well as teaching and academic coordination. I’ve also contributed AI training data at Mercor by recording and narrating computer workflows.

