My (Chiffon) Nguyen she/her
Nguyễn Trà My / 阮沐茶 / 윈자미
AI Researcher & Life-long Learner
Pronunciation
My = /me/, Chiffon = /shi-FON/. I'm Chiffon in English-speaking context.
I research how to ensure that current and future AI systems reflect and support pluralism across languages, cultures, and social contexts.
- Pluralistic alignment, and more broadly AI Safety
- Multilingual & multicultural AI
- AI & Society: education, information & media, human-AI collaboration
I care about AI ethics, and impacts of AI on humans, society, and more-than-human world↗.
Ongoing work: CoT monitoring (Algoverse), multilingual agentic eval (SEACrowd), extension to AnimalHarmBench (ElectricSheep), and multicultural benchmarking (Cohere Labs).
I am applying to research-oriented Master’s in Computer Science, AI, NLP, or related fields for Fall 2026/Spring 2027 and seeking research opportunities (internships, research assistantships, collaborations).
Career highlights
see CV →- - Present
AI Research Mentee (Multilingual Agentic Evaluation) @ SEACrowd · Remote
Extending Tau3-Bench for multilingual agentic evaluation across 5 settings. Mentors: Dr. Samuel Cahyawijaya (Cohere) & Patomporn Payoungkhamdee (VISTEC). Target: EMNLP '26.
- - Present
AI Research Mentee (Chain-of-thought Monitoring) @ Algoverse AI Research · Remote
Lead project investigating whether weaker models can detect sandbagging in stronger models on safety evals. Evaluated 20 open-weight reasoning model pairs. Target: ICML '26 Workshop
- -
B.Sc in Computational Sciences @ Minerva University · San Francisco, CA, USA (+ Seoul, Taipei, Hyderabad, Buenos Aires, Berlin)
GPA 3.67/4.00. Relevant Coursework: Machine Learning, Bayesian Modeling, Statistical Modeling & Causal Inference, Optimization Methods.
- -
ML Research Assistant @ Landshut University of Applied Sciences · Landshut, Bavaria, Germany
3D object detection in point clouds and point cloud registration, supervised by Prof. Eduard Kromer and Prof Sandra Eisenreich
Latest updates
see all →Started Futurekind AI Fellowship Spring 2026↗ where I research and build solutions in the intersection of AI and animal protection.
CommonLID↗ got accepted to ACL 2026 Main Conference.
[Till Jun 2026] Working on extending Tau3-Bench for low-resource languages as a mentee at SEACrowd 2026 Apprentice Program↗.
Miscellaneous
-
My Vietnamese name is ‘Trà My’, which means Camellia japonica↗ (tea flower).
-
I graduated in May 2025 from Minerva University↗ studying Machine Learning & Statistics, with peers from 50+ countries and living in 6 cities as part of its global immersion program↗: Seoul (South Korea), Chinese Taipei, Hyderabad (India), Buenos Aires (Argentina), Berlin (Germany). This experience heavily shapes my worldviews and motivates my research focus in diversity and collaboration.
-
Things that restore my energy: walks, reading↗, cooking, cycling, card reading, and astrology. I also enjoy solo travel, food, cultural activities, and historical museums.
-
Things that excite me outside research: teaching & mentoring, event planning and organizing, languages and cultures, philosophical and politcal discussions.
-
I’m learning Mandarin Chinese (band 4/B1) and French (A2).
-
I’m agnostic, with wonderful friends that are Christian↗, Buddhist↗, and Muslim↗.
-
I’m on no-meat diet at least 4 days/week, and trying to transition to vegetarian diet.
-
I like Rust-based and well-designed stuff. Recommend: Astral’s Python tooling↗ (uv, ruff, ty), Inspect↗ for reproducible LLM eval, LMMs-eval↗ for multimodal eval, Rust-based CLI (ripgrep↗, eza↗, bat↗). My hardware and apps here.