Quy Nguyen

Designed and evaluated LLM-based solutions for translation quality assessment, multilingual voice assistant QA, and brand-compliant content checks. Applied GPT-4o, RAG, and LLM-as-a-judge methods across automotive domains. Improved translation memories and corporate terminology by integrating market feedback and leveraging LLMs to detect linguistic, terminological, and stylistic inconsistencies in technical content.

Investigated fairness and bias in NLG systems and evaluated metrics for assessing fairness in large language models. Contributed to internal research seminars on responsible AI and ethical use of LLMs.

Constructed annotated corpora and conducted linguistic data analysis for Chinese and Vietnamese. Published two first-author papers on contrastive syntax between the two languages.