PeopleofSuperintelligence

Researcher

Heng-Tze Cheng

Heng-Tze Cheng is a Google engineer whose work spans recommender systems, language models and Gemini 1.0. He earned a bachelor's degree in electrical engineering at National Taiwan University in 2008 and a PhD in electrical and computer engineering at Carnegie Mellon University in 2013. He joined Google in 2014, and he founded and led the Wide & Deep Learning project in TensorFlow. He co-authored LaMDA: Language Models for Dialog Applications, and in 2023 he served as co-lead of Gemini App Research on Gemini 1.0. As of October 2026, he is a Technical Lead Manager and Senior Staff Software Engineer on the Google Brain team, part of Google Research & AI.

Works
39
Citations
6,623
h-index
19

Publication numbers from OpenAlex, as of 15 October 2026.

Timeline

  1. 2008–2013

    Electrical engineering degrees

    He earns a bachelor's degree in electrical engineering at National Taiwan University in 2008 and a PhD in electrical and computer engineering at Carnegie Mellon University in 2013.

  2. 2014

    Joining Google

    He joins Google in 2014, after earning his PhD at Carnegie Mellon University in 2013.

  3. 2016

    The Wide & Deep Learning project

    He co-writes Wide & Deep Learning for Recommender Systems in 2016. He founds and leads the Wide & Deep Learning project in TensorFlow.

  4. 2022

    The LaMDA language model

    He co-authors LaMDA: Language Models for Dialog Applications, a paper submitted to arXiv in January 2022.

  5. 2023

    Gemini app research co-lead

    He co-leads Gemini App Research on Gemini 1.0, a report published in December 2023.

  6. 2026

    Autonomous mathematics research

    He co-authors Towards Autonomous Mathematics Research in 2026, a paper whose work was conducted under Google DeepMind.

  7. As of October 2026

    Technical lead at Google Research

    He is a Technical Lead Manager and Senior Staff Software Engineer on the Google Brain team, part of Google Research & AI. He currently leads a research team on neural sequence modeling.

Sources and credits

Photo: We have no published photo of Heng-Tze Cheng yet. Send us a free-licensed photo.

  1. Heng-Tze Cheng users.ece.cmu.edu. Checked 15 October 2026.
  2. Heng-Tze Cheng research.google. Checked 15 October 2026.
  3. [1606.07792] Wide & Deep Learning for Recommender Systems arxiv.org. Checked 15 October 2026.
  4. [2201.08239] LaMDA: Language Models for Dialog Applications arxiv.org. Checked 15 October 2026.
  5. Gemini: A Family of Highly Capable Multimodal Models arxiv.org. Checked 15 October 2026.
  6. Gemini: A Family of Highly Capable Multimodal Models arxiv.org. Checked 15 October 2026.
  7. Towards Autonomous Mathematics Research arxiv.org. Checked 15 October 2026.