My Chiffon N.
My (Chiffon) Nguyen

My (Chiffon) Nguyen she/her

Nguyễn Trà My / 阮沐茶 / 윈자미

AI Research for Broader World & Life-long Learning

Pronunciation

My = /me/, Chiffon = /shi-FON/. I'm Chiffon in English-speaking context.

I research current and future AI for human plurality and broader world. Towards this end, I’m currently interested in the following problems:

For technical research, I’m interested in data work and evaluation to make grounded, predictive, specific claims about long-horizon AI capability and safety, then improve them.

I’m interning at Lida Safety Research. I also contribute to community research at various capacities, including Cohere Labs Community (leadership & analysis), MatrAIx (AI simulation apps), and BenchFlow (evaluation infrastructure). Recently I worked on SEATauBench, and CoT monitoring behavior.

I am seeking industry roles in SWE & AI Engineering and research-oriented Master’s in Computer Science, AI, NLP, or related fields for Fall 2027 or later. For research collaborations, drop me an email!

Career highlights

see CV →

Latest updates

see all →

Started interning at Lida Safety Research, contributing to projects in mech interp and sandbagging.

Started co-leading Cohere Labs Open Science Community’s multi-cultural riddles benchmarking project with other six leads, working with 100+ community researchers.

Selected publications

see all →

SEATauBench: Adapting Tool-Agent-User Evaluation Into Low-Resource Southeast Asian Languages
My Chiffon Nguyen*, Aulia Adila*, Saksorn Ruangtanusak*, Kittiphat Leesombatwathana*, Vissuta Gunawan Lim*, Patomporn Payoungkhamdee, Samuel Cahyawijaya,
In Submission • 2026
Abstract
We introduce SEATauBench, the first agentic-focused evaluation framework for sovereign AI development in Southeast Asia, a region of strategic importance with over 700 million people. Despite growing regional evaluation efforts, existing multilingual agents show limited capability when operating in SEA languages, particularly in mixed-language scenarios. Through evaluation across multiple adaptation approaches, we find that while English agentic capabilities transfer to target language responses, performance degrades significantly when context is provided in SEA languages. We propose a translation-based mitigation strategy that preserves entity consistency while enabling agents to leverage English comprehension. SEATauBench establishes a rigorous benchmark for sovereign AI agent assessment, providing diagnostic tools to address capability gaps and support agentic AI development in diverse linguistic communities in the region.
PDF
Evaluation/Benchmarking Agent Multilinguality

Miscellaneous

  • My Vietnamese name is ‘Trà My’, which is 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), Taipei, Hyderabad (India), Buenos Aires (Argentina), Berlin (Germany). This experience heavily shapes my worldviews and motivates my research focus in diversity and collaboration.

  • In no order: reading, cooking, cycling, travel, teaching, language learning, cultural activities, historical museums, event organizing, philosophy, and politics.

  • My Mandarin Chinese (band 4/B1) is for shows and French (A2) for memes.

  • I’m agnostic, with wonderful friends that are Christian, Buddhist, and Muslim.

  • My friends said I have a lot of good (& niche) tech recommendation, and ask me to collect them in a page, so I have one here.