PeopleofSuperintelligence

Builder

Costa Huang

Also known as Shengyi Huang

Costa Huang is a machine-learning engineer whose work spans deep reinforcement learning and open language models. He created CleanRL, a library for game-playing agents, and coauthored a 2022 paper on its single-file algorithms. He completed a computer-science PhD at Drexel University in 2023 with a dissertation on deep reinforcement learning. He was a researcher and machine-learning engineer at Ai2 in 2024; he coauthored Olmo 3, a family of fully open 7B and 32B language models, in 2025. As of October 2026, he is a team member at Periodic Labs.

Works
67

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

    Developing CleanRL

    In 2019, he develops CleanRL, a deep-reinforcement-learning library for training game-playing agents.

  2. 2022

    A CleanRL journal paper

    In 2022, he coauthors a Journal of Machine Learning Research paper describing CleanRL's single-file deep reinforcement-learning algorithms.

  3. 2023

    A Drexel doctorate

    In 2023, he completes a computer-science PhD at Drexel University with a dissertation titled Reproducible and Efficient Deep Reinforcement Learning.

  4. 2024

    Research engineering at Ai2

    In 2024, he is a researcher and machine-learning engineer at Ai2.

  5. 2025

    Coauthoring Olmo 3

    In 2025, he coauthors Olmo 3, a family of fully open language models at the 7B and 32B parameter scales.

  6. As of October 2026

    On Periodic Labs' team

    As of October 2026, he is a member of Periodic Labs' team.

Career

  1. NowPeriodic LabsTeam Member
  2. From 2024Allen Institute for AI (Ai2)Researcher and Machine-Learning Engineer

Education

  1. 2023Drexel UniversityPhD

From the biography.

  1. Google alumni founded or led 87 other organizations, the most among the labs we compareIncludes Periodic Labs. We trace 212 organizations founded or led by AI lab alumni, and 223 career moves between the labs.
  2. The company share of AI researchers’ top papers rose from 2% to 32%Includes Allen Institute for AI (Ai2). We estimate where leading AI researchers worked when their most-cited papers appeared. The company share rose from 2% in 1975–1979 to 32% in 2020–2024.
  3. Google has the highest combined h-index among AI organizationsIncludes Allen Institute for AI (Ai2). Our estimates of combined and median h-index at AI organizations, using current affiliations and OpenAlex records.
  4. 44% of AI career moves go from one university to anotherIncludes Allen Institute for AI (Ai2). 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 Costa Huang yet. Send us a free-licensed photo.

  1. Costa Huang's Website costa.sh. Checked 19 October 2026.
  2. CleanRL: High-quality Single-file Implementations of Deep Reinforcement Learning Algorithms jmlr.org. Checked 19 October 2026.
  3. Santiago Ontañón Villar sites.google.com. Checked 19 October 2026.
  4. 2022-23 Doctoral Defenses | Office of Graduate Studies | Drexel University drexel.edu. Checked 19 October 2026.
  5. 'Reasoning' AI models have become a trend, for better or worse techcrunch.com. Checked 19 October 2026.
  6. Olmo 3 arxiv.org. Checked 19 October 2026.
  7. Periodic Labs periodic.com. Checked 19 October 2026.