PeopleofSuperintelligence

Researcher

Jingchang Chen

Jingchang Chen is one of the developers of DeepSeek Elastic Compute (DSec), the sandbox platform that DeepSeek-AI uses for agentic training at scale. With Zheng Chu and colleagues, they coauthored a survey of chain-of-thought reasoning that appeared at ACL 2024. At Harbin Institute of Technology, they were first author of FunCoder, a code-generation framework that NeurIPS 2024 accepted as an oral presentation. They coauthored DeepSeek's reports on DeepSeek-V3 in December 2024, DeepSeek-R1 in January 2025, DeepSeek-V3.2 in December 2025 and DeepSeek-V4 in April 2026, and Nature published the R1 study in September 2025.

Now at
DeepSeek

Timeline

  1. 2023–2024

    A chain-of-thought reasoning survey

    With Zheng Chu, Qianglong Chen, Ming Liu, Bing Qin and others, they coauthor a survey of chain-of-thought reasoning in large language models, first posted in September 2023. The paper appears at ACL 2024 in Bangkok.

  2. 2024

    FunCoder at NeurIPS 2024

    At Harbin Institute of Technology, they are first author of the 2024 FunCoder paper, written with Hongxuan Tang, Zheng Chu, Ming Liu, Bing Qin and others. FunCoder generates code by recursively branching off sub-functions and choosing among candidate functions through a consensus on their behavior. NeurIPS 2024 in Vancouver accepts the paper as an oral presentation.

  3. December 2024

    The DeepSeek-V3 technical report

    They are among the authors of the DeepSeek-V3 technical report, posted in December 2024. DeepSeek-V3 is a Mixture-of-Experts language model with 671 billion total parameters, 37 billion of them activated for each token, and its full training takes 2.788 million H800 GPU hours.

  4. 2025

    DeepSeek-R1 in Nature

    In January 2025, they coauthor DeepSeek-R1, which shows that the reasoning abilities of large language models can be incentivized through pure reinforcement learning. Nature publishes the study in September 2025, with its authors at the DeepSeek-AI Team in Hangzhou, China.

  5. December 2025

    DeepSeek-V3.2 and sparse attention

    They coauthor the DeepSeek-V3.2 report in December 2025. The model introduces DeepSeek Sparse Attention, and its high-compute variant, DeepSeek-V3.2-Speciale, achieves gold-medal performance at the 2025 International Mathematical Olympiad and the International Olympiad in Informatics.

  6. April 2026

    The DeepSeek-V4 preview

    In April 2026, they coauthor the preview report on the DeepSeek-V4 series, in which DeepSeek-V4-Pro has 1.6 trillion parameters and DeepSeek-V4-Flash has 284 billion. Both models support a context length of one million tokens.

  7. September 2026

    The DeepSeek-V4.1-Flash model

    In September 2026, they coauthor the report on DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model with 552 billion backbone parameters and contexts of up to one million tokens. The model reduces its global KV cache footprint to 890 bytes per token.

  8. As of October 2026

    Developing DSec at DeepSeek

    They are one of the DSec project developers at DeepSeek-AI, alongside Hongxuan Tang. DeepSeek Elastic Compute (DSec), presented in a September 2026 report, is a production sandbox platform for agentic training, and a single unit serves about 3 million sandboxes per day.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes DeepSeek. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes DeepSeek. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.
  3. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staffIncludes DeepSeek. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staff. We test the estimate against 3,913 named model contributors and dated all-staff records across Google DeepMind, OpenAI and Anthropic, plus six additional frontier-grade labs.
  4. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes DeepSeek. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.

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 Jingchang Chen yet. Send us a free-licensed photo.

  1. Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Future (arXiv:2309.15402) arxiv.org. Checked 20 October 2026.
  2. Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Future - ACL Anthology aclanthology.org. Checked 20 October 2026.
  3. Divide-and-Conquer Meets Consensus: Unleashing the Power of Functions in Code Generation (arXiv:2405.20092) arxiv.org. Checked 20 October 2026.
  4. Divide-and-Conquer Meets Consensus: Unleashing the Power of Functions in Code Generation - Harbin Institute of Technology hit.globalimpact.cn. Checked 20 October 2026.
  5. NeurIPS Oral Divide-and-Conquer Meets Consensus: Unleashing the Power of Functions in Code Generation neurips.cc. Checked 20 October 2026.
  6. DeepSeek-V3 Technical Report (arXiv:2412.19437) arxiv.org. Checked 20 October 2026.
  7. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning (arXiv:2501.12948) arxiv.org. Checked 20 October 2026.
  8. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning | Nature nature.com. Checked 20 October 2026.
  9. DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models (arXiv:2512.02556) arxiv.org. Checked 20 October 2026.
  10. DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence (arXiv:2606.19348) arxiv.org. Checked 20 October 2026.
  11. DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression (arXiv:2609.19969) arxiv.org. Checked 20 October 2026.
  12. DeepSeek Elastic Compute (DSec): A Sandbox Infrastructure for Effective Agentic Training at Scale (arXiv:2609.22978) arxiv.org. Checked 20 October 2026.
  13. DeepSeek Elastic Compute (DSec): A Sandbox Infrastructure for Effective Agentic Training at Scale (HTML) arxiv.org. Checked 20 October 2026.