Arya Desai

I am an AI Evaluator and Data Annotator with hands-on experience in LLM evaluation, RLHF pipelines, and structured data annotation across text, code, and
image modalities.At Turing I evaluated 150+ production-grade LLM outputs weekly using structured rubrics and adversarial test scenarios for a top-tier global technology client. I currently deliver annotation and evaluation work across multiple AI platforms, consistently passing platform quality checks across structured sprint cycles.I bring strong analytical judgment, attention to detail, and the ability to follow evolving guidelines with consistency across high volume tasks. I am
native in Gujarati, professional in English (IELTS 7.0), and conversational in Hindi, making me well suited for multilingual annotation projects.

Services I offer: LLM output evaluation, rubric-based annotation, failure mode classification, preference ranking, adversarial prompt testing, multilingual content evaluation, and data quality review across AI training pipelines.

Available immediately for freelance and contract engagements. Flexible hours, remote only.