PeopleofSuperintelligence

Researcher

Hongyu Chen

Also known as Charlie Chen

Hongyu Chen, also known as Charlie Chen, is a Cohere machine-learning engineer whose work spans language models, model compression, and AI safety. She starts with AI reporting at Synced, studies neural-network quantization at Qualcomm, and researches machine-learning security under Nicolas Papernot at the University of Toronto. Her work includes a 2023 NeurIPS study on quantization at scale, a 2025 study of artifacts in LLM safety evaluations, and Cohere's Command A report. In 2026, she mentors a Cohere Labs project that tests multilingual safety with mixed-language prompts.

Timeline

  1. May 2020–August 2020

    AI reporting at Synced

    She prepares reports on AI technology trends as a business intelligence analyst at Synced from May to August 2020.

  2. September 2020–September 2021

    Quantization research at Qualcomm

    At Qualcomm, she works as a Machine Learning Research Engineer from September 2020 to September 2021, studying neural-network quantization and AutoML.

  3. May 2021

    Student researcher at CleverHans

    From May 2021, she works as a Student Researcher with the University of Toronto's CleverHans Lab under Nicolas Papernot, completing an undergraduate thesis and collaborating on machine-learning security.

  4. 2022

    A Toronto machine-intelligence degree

    She earns a Bachelor of Applied Science in Machine Intelligence at the University of Toronto in 2022.

  5. 2023

    Quantization across model scales

    In 2023, she coauthors Cohere's NeurIPS study “Intriguing Properties of Quantization at Scale” under the name Hongyu (Charlie) Chen. The study reports quantizing models from 410M to 52B with minimal performance degradation.

  6. 2025

    Testing LLM safety evaluators

    In 2025, she and Seraphina Goldfarb-Tarrant study LLM safety evaluators' susceptibility to input artifacts. They find that apologetic language can skew evaluator preferences by up to 98%.

  7. 2025

    The Command A model report

    In 2025, she coauthors Cohere's technical report “Command A: An Enterprise-Ready Large Language Model.”

  8. As of October 2026

    Cohere Labs mentorship

    She is a Machine Learning Engineer at Cohere. In 2026, she mentors Cohere Labs' “Mix, Fine-Tune, Break” project, which tests multilingual safety with mixed-language prompts.

  1. Google had the most people shaping AI every year from 2014 to 2025Our estimates of which organizations employed the most people shaping AI from 1961 to 2025, with a year-by-year chart.
  2. The United States employs an estimated 53% of people in our data with a known employment countryOur estimates of where AI careers are concentrated now and how countries’ shares of dated positions changed over time.
  3. 46% of living people in our data with known birth and employment countries work outside their birth countryOur estimate of cross-border careers in AI, from country of birth and the country of each current organization.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionOur estimates of where the people shaping AI studied and where their careers led, from education records in cited biographies.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Hongyu Chen yet. Send us a free-licensed photo.

  1. Academic hongyu-charlie.netlify.app. Checked 19 October 2026.
  2. Intriguing Properties of Quantization at Scale cohere.com. Checked 19 October 2026.
  3. Intriguing Properties of Quantization at Scale papers.nips.cc. Checked 19 October 2026.
  4. Safer or Luckier? LLMs as Safety Evaluators Are Not Robust to Artifacts - ACL Anthology aclanthology.org. Checked 19 October 2026.
  5. [2504.00698] Command A: An Enterprise-Ready Large Language Model arxiv.org. Checked 19 October 2026.
  6. CleverHans Lab - Members cleverhans.io. Checked 19 October 2026.
  7. Tiny Aya Expedition Drives Multilingual Innovation | Cohere cohere.com. Checked 19 October 2026.