I am a research and data professional with more than three years of experience supporting AI and data-focused work. My background includes LLM evaluation, data verification, prompt testing, research, and quality review.
I work with structured and text-based datasets, focusing on accuracy, consistency, and reliable source validation. I use Python and SQL to clean, filter, organize, and validate data for analytical and model-training workflows.
I have experience evaluating LLM responses for factuality, relevance, safety, logical coherence, and alignment. I also conduct response ranking, data labeling, bias testing, hallucination detection, and multi-turn conversation reviews.
I am skilled in prompt engineering and adversarial testing, using edge cases and challenging scenarios to identify model failure modes and evaluate compliance with system constraints. I translate unclear stakeholder requirements into practical specifications and documented workflows.
My research experience includes primary-source verification, fact-checking, literature review, and technical documentation. I communicate findings clearly through analytical reports, SOPs, internal style guides, and workflow materials.
I bring an analytical, detail-oriented approach to cross-functional collaboration, with a strong commitment to data quality, AI safety, and continuous process improvement.
