PeopleofSuperintelligence

Researcher

Christopher Hesse

Christopher Hesse is an OpenAI technical staff member whose AI work spans reinforcement learning, language models, code generation, and mathematical reasoning. His early OpenAI work included the Dota 2 study behind OpenAI Five and the Procgen benchmark. He co-authored papers on GPT-3 few-shot learning, Codex and HumanEval, verifiers for grade-school math, and WebGPT's browser-assisted answers with human feedback. He later co-authored OpenAI's GPT-4o and o1 system cards. As of October 2026, he is a Member of Technical Staff at OpenAI.

Now at
OpenAI
Works
11

Publication numbers from OpenAlex, as of 19 October 2026. Citations and h-index are hidden because this OpenAlex record may be incomplete or mix in other authors.

Timeline

  1. 2019

    OpenAI Five and Procgen

    In 2019, he co-authors OpenAI's Dota 2 study, which describes OpenAI Five's win over world champion Team OG. Also in 2019, he co-authors Procgen Benchmark, a suite of 16 procedurally generated game-like environments for measuring sample efficiency and generalization in reinforcement learning.

  2. 2020

    GPT-3 few-shot learning

    In 2020, he co-authors “Language Models are Few-Shot Learners,” which tests a 175-billion-parameter GPT-3 on tasks specified through text interaction and few-shot demonstrations.

  3. 2021

    Codex and HumanEval

    In 2021, he co-authors the Codex study, which introduces HumanEval to measure functional correctness for programs synthesized from docstrings. OpenAI's repository provides an evaluation harness for the dataset.

  4. 2021

    Math verifiers and GSM8K

    In 2021, he co-authors research that introduces GSM8K, a dataset of 8.5K grade-school math problems, and trains verifiers to rank candidate solutions.

  5. 2021

    WebGPT and human feedback

    In 2021, he co-authors WebGPT, which fine-tunes GPT-3 to answer long-form questions in a text-based web-browsing environment and improves answer quality with human feedback.

  6. 2024

    GPT-4o system card

    In 2024, he co-authors OpenAI's GPT-4o System Card, which details its capabilities, limitations, and safety evaluations across speech, text, and images.

  7. 2024–2026

    The o1 system card

    In 2024, he co-authors OpenAI's o1 System Card, which outlines safety evaluations and external red teaming. The team revises the report in April 2026.

  8. Since 2016

    OpenAI technical staff

    As of October 2026, he is a Member of Technical Staff at OpenAI.

  1. 13 of 79 people shaping AI with a former OpenAI role now work at AnthropicIncludes OpenAI. Our estimates of where OpenAI recruits came from and where people with former OpenAI roles work now.
  2. The median h-index is 53 at universities and 19 at frontier AI labsIncludes OpenAI. The median h-index is 53 at universities and 19 at frontier AI labs. We compare publishing careers, citations and moves to estimate how universities and frontier labs contribute to AI research.
  3. Google alumni founded or led 87 other organizations, the most among the labs we compareIncludes OpenAI. We trace 212 organizations founded or led by AI lab alumni, and 223 career moves between the labs.
  4. 44% of AI career moves go from one university to anotherIncludes OpenAI. We estimate how 14,272 AI career moves link universities, companies, labs and government research centers, with net flows and the most common paths.

Colleagues

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

Sources and credits

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

  1. Dota 2 with Large Scale Deep Reinforcement Learning arxiv.org. Checked 19 October 2026.
  2. Leveraging Procedural Generation to Benchmark Reinforcement Learning arxiv.org. Checked 19 October 2026.
  3. Language Models are Few-Shot Learners arxiv.org. Checked 19 October 2026.
  4. Evaluating Large Language Models Trained on Code arxiv.org. Checked 19 October 2026.
  5. GitHub - openai/human-eval: Code for the paper Evaluating Large Language Models Trained on Code github.com. Checked 19 October 2026.
  6. Training Verifiers to Solve Math Word Problems arxiv.org. Checked 19 October 2026.
  7. WebGPT: Browser-assisted question-answering with human feedback arxiv.org. Checked 19 October 2026.
  8. GPT-4o System Card arxiv.org. Checked 19 October 2026.
  9. OpenAI o1 System Card arxiv.org. Checked 19 October 2026.
  10. Christopher Hesse | alphaXiv alphaxiv.org. Checked 19 October 2026.