PeopleofSuperintelligence

Researcher

Fangyun Lin

Fangyun Lin is an AI researcher whose work with DeepSeek spans language models, reasoning and inference. They coauthored the 2024 DeepSeek LLM, DeepSeek-V2 and DeepSeek-V3 reports, including work on open-source model scaling and mixture-of-experts designs. In 2025, they coauthored DeepSeek-R1's reinforcement-learning research and DeepSeek-V3.2's sparse-attention work. As of October 2026, they are on DeepSeek's Research & Engineering team and coauthor DeepSeek-V4.1-Flash.

Timeline

  1. January 2024

    Open-source language models

    They coauthor the DeepSeek LLM report in January 2024, which uses scaling laws to study open-source 7B and 67B language models.

  2. May 2024

    DeepSeek-V2's mixture-of-experts model

    In May 2024, Lin coauthors DeepSeek-V2, a mixture-of-experts model with 236 billion total parameters and 21 billion activated per token.

  3. December 2024

    DeepSeek-V3's technical report

    The December 2024 DeepSeek-V3 report describes a 671-billion-parameter mixture-of-experts model with 37 billion active per token; Lin coauthors it.

  4. 2025

    DeepSeek-R1 and reinforcement learning

    In 2025, they coauthor the DeepSeek-R1 paper, which shows that pure reinforcement learning can elicit reasoning without human-labelled trajectories.

  5. December 2025

    DeepSeek Sparse Attention

    The DeepSeek-V3.2 report introduces DeepSeek Sparse Attention in December 2025, a mechanism that reduces computational complexity in long-context scenarios; Lin coauthors the report.

  6. As of October 2026

    Research and engineering at DeepSeek

    They work in DeepSeek's Research & Engineering team as of October 2026 and coauthor the group's V4.1-Flash report on KV-cache compression. The report describes a multimodal mixture-of-experts model with a Causal Encoder-Decoder architecture and up to one million tokens of context.

  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.

Sources and credits

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

  1. [2401.02954] DeepSeek LLM: Scaling Open-Source Language Models with Longtermism arxiv.org. Checked 19 October 2026.
  2. [2405.04434] DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model arxiv.org. Checked 19 October 2026.
  3. DeepSeek-V3 Technical Report | Cool Papers - Immersive Paper Discovery papers.cool. Checked 19 October 2026.
  4. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning | Nature nature.com. Checked 19 October 2026.
  5. [2512.02556] DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models ar5iv.labs.arxiv.org. Checked 19 October 2026.
  6. [2609.19969] DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression ar5iv.labs.arxiv.org. Checked 19 October 2026.