PeopleofSuperintelligence

Researcher

Luyu Gao

Luyu Gao is an AI researcher whose work spans dense information retrieval, language-model reasoning, and scientific coding. He studied computer science at the University of Illinois Urbana-Champaign, completed a six-month internship with NVIDIA's GPU Foundation Profiler team, and earned a Master of Language Technologies at Carnegie Mellon in 2021. His research includes Condenser for dense retrieval, PAL's program-aided reasoning, Self-Refine's feedback loop, RARR's attribution and revision, and the SciCode coding benchmark. In June 2025, he coauthored the Magistral paper on Mistral AI's first reasoning model; as of October 2026, he is a machine learning engineer at OpenAI working on neural information retrieval.

Now at
OpenAI
Badges

Timeline

  1. 2021

    Condenser for dense retrieval

    In 2021, he coauthors Condenser, a pre-training Transformer architecture that improves over standard language models on text retrieval and similarity tasks.

  2. By 2022

    UIUC, NVIDIA and Carnegie Mellon

    He graduates from the University of Illinois Urbana-Champaign with a computer science major. In 2021, he earns a Master of Language Technologies at Carnegie Mellon's Language Technologies Institute, with Jamie Callan as academic advisor. By 2022, he has completed a six-month internship with NVIDIA's GPU Foundation Profiler team.

  3. 2022–2023

    PAL for language model reasoning

    In 2022, he and his coauthors post PAL as a preprint; in 2023, the work appears in the Proceedings of Machine Learning Research at ICML. The method has a language model generate programs as intermediate reasoning steps and sends them to a Python interpreter for execution.

  4. 2023

    Self-Refine's feedback loop

    In 2023, he coauthors Self-Refine, which uses one language model to generate an output, give feedback on it, and revise it iteratively.

  5. 2023

    RARR for generated text

    In 2023, he coauthors RARR, a system that finds attribution for generated text and edits unsupported content while preserving the original output as much as possible.

  6. 2024

    A benchmark for scientific coding

    In 2024, he coauthors SciCode, a coding benchmark created with input from scientists and AI researchers across 16 natural-science fields. It evaluates language models on code for real scientific research problems.

  7. June 2025

    Mistral's first reasoning model

    In June 2025, he coauthors the Magistral paper. It introduces Mistral AI's first reasoning model and describes a scalable reinforcement-learning pipeline.

  8. As of October 2026

    OpenAI machine learning engineer

    As of October 2026, he is a machine learning engineer at OpenAI and works on neural information retrieval.

Career

  1. NowOpenAIMachine Learning Engineer
  2. NVIDIAIntern

Education

  1. 2021Carnegie Mellon UniversityMaster of Language Technologies
  2. University of Illinois Urbana-Champaign

From the biography.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Luyu Gao. 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 Luyu Gao. 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. Among employers with 30+ full-time cases, OpenAI leads primary technical role H-1B pay at $310,000, and Amazon, Google and Microsoft file 57% of all casesIncludes OpenAI. Across 440,669 certified US H-1B filings from 44 employers (FY2020–FY2026), OpenAI leads primary technical role base pay at $310,000 and xAI leads overall base pay at $290,000.
  4. 22 of 133 people shaping AI with a former OpenAI role now work at AnthropicIncludes OpenAI. Our estimates of where OpenAI recruits came from and where 133 people with former OpenAI roles work now, including the companies and labs they founded.

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 Luyu Gao yet. Send us a free-licensed photo.

  1. Condenser: a Pre-training Architecture for Dense Retrieval aclanthology.org. Checked 21 October 2026.
  2. Luyu Gao luyug.github.io. Checked 21 October 2026.
  3. Luyu Gao - Language Technologies Institute - School of Computer Science - Carnegie Mellon University lti.cmu.edu. Checked 21 October 2026.
  4. [2211.10435] PAL: Program-aided Language Models arxiv.org. Checked 21 October 2026.
  5. PAL: Program-aided Language Models proceedings.mlr.press. Checked 21 October 2026.
  6. Self-Refine: Iterative Refinement with Self-Feedback proceedings.neurips.cc. Checked 21 October 2026.
  7. RARR: Researching and Revising What Language Models Say, Using Language Models aclanthology.org. Checked 21 October 2026.
  8. [2407.13168] SciCode: A Research Coding Benchmark Curated by Scientists arxiv.org. Checked 21 October 2026.
  9. [2506.10910] Magistral arxiv.org. Checked 21 October 2026.
  10. Jamie Callan's Research Group cs.cmu.edu. Checked 21 October 2026.