PeopleofSuperintelligence

Researcher

Qihao Zhu

Also known as 朱琪豪

Qihao Zhu is an AI researcher whose work carries structured code generation and repair into large language models. He coauthored TreeGen, Recoder and Tare, and his doctoral thesis won a 2024 Distinguished Thesis Award from TCSE and CCF. At DeepSeek, he coauthored DeepSeek-Coder-V2 and DeepSeek-V3, was named a core contributor to DeepSeek-R1, and coauthored DeepSeek-Prover-V2. As of October 2026, he is a researcher at DeepSeek in Beijing, working on code and formal reasoning.

Now at
DeepSeek
Works
52
Citations
2,305
h-index
20
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2020

    TreeGen for code generation

    In 2020, he coauthors TreeGen, a tree-based neural architecture for code generation. It uses Transformer attention and an AST reader to incorporate grammar rules and syntax-tree structure.

  2. August 2021–May 2023

    Syntax-guided code repair

    In August 2021, he coauthors A Syntax-Guided Edit Decoder for Neural Program Repair for ESEC/FSE. In May 2023, he coauthors Tare for ICSE, a type-aware model that integrates type information into a grammar and guides patch generation with typing rules.

  3. June 2024

    DeepSeek-Coder-V2 for code

    In June 2024, he coauthors DeepSeek-Coder-V2, an open-source mixture-of-experts code model. It expands support from 86 to 338 programming languages.

  4. December 2024

    A distinguished thesis

    In December 2024, his doctoral thesis wins the Distinguished Thesis Award from TCSE, CCF. Yingfei Xiong and Lu Zhang co-advise the thesis, which guides neural networks to learn programming-language syntax and types for defect repair and code search.

  5. December 2024

    Coauthoring DeepSeek-V3

    In December 2024, he coauthors the DeepSeek-V3 technical report, which presents a mixture-of-experts language model with 671 billion total parameters and 37 billion activated per token.

  6. January 2025

    Reasoning in DeepSeek-R1

    In January 2025, he is named among DeepSeek-R1's core contributors. The report describes reasoning models trained through reinforcement learning.

  7. April 2025

    Formal theorem proving

    In April 2025, he coauthors DeepSeek-Prover-V2, an open-source model for formal theorem proving in Lean 4. Its training pipeline uses DeepSeek-V3 to decompose complex problems into subgoals before reinforcement learning.

  8. As of October 2026

    Researcher at DeepSeek

    As of October 2026, he is a researcher at DeepSeek in Beijing. His work covers code language models, mathematical and formal reasoning, and reinforcement learning.

Career

  1. Since 2024DeepSeekResearcher
  2. 2022–2025Singapore University of Technology and Design
  3. 2018–2025Peking University

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. Deregulated microRNAs in gastric cancer tissue-derived mesenchymal stem cells: novel biomarkers and a mechanism for gastric cancer2014 · British Journal of Cancer · 230 citations
  2. A syntax-guided edit decoder for neural program repair2021 · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · 219 citations
  3. The IL-6–STAT3 axis mediates a reciprocal crosstalk between cancer-derived mesenchymal stem cells and neutrophils to synergistically prompt gastric cancer progression2014 · Cell Death and Disease · 189 citations
  4. DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence2024 · 179 citationsWith Daya Guo
  5. TreeGen: A Tree-Based Transformer Architecture for Code Generation2020 · Proceedings of the AAAI Conference on Artificial Intelligence · 171 citations

From OpenAlex, as of 21 October 2026.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Qihao Zhu. 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. 16% of AI career moves in the 2020s went from a university to a companyIncludes Qihao Zhu. We estimate that 16% of AI career moves in the 2020s went from a university to a company; 29% of moves starting at a university went to a company, and 76% ended the university role.
  3. 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.
  4. 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.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

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

  1. TreeGen: A Tree-Based Transformer Architecture for Code Generation ojs.aaai.org. Checked 20 October 2026.
  2. A Syntax-Guided Edit Decoder for Neural Program Repair · Yingfei Xiong xiongyingfei.github.io. Checked 20 October 2026.
  3. Tare: Type-Aware Neural Program Repair (ICSE 2023 - Technical Track) - ICSE 2023 conf.researchr.org. Checked 20 October 2026.
  4. DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence arxiv.org. Checked 20 October 2026.
  5. Ph.D. Thesis of Qihao Zhu won the Distinguished Thesis Award by TCSE, CCF-Programming Languages Lab, Peking University pl.cs.pku.edu.cn. Checked 20 October 2026.
  6. DeepSeek-V3 Technical Report arxiv.org. Checked 20 October 2026.
  7. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning arxiv.org. Checked 20 October 2026.
  8. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning arxiv.org. Checked 20 October 2026.
  9. DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition arxiv.org. Checked 20 October 2026.
  10. Qihao Zhu 朱琪豪 · Code Intelligence & Reasoning · DeepSeek pkuzqh.github.io. Checked 20 October 2026.