Samuel Wang

AI and Robotics @ UPenn | Ai2

Philadelphia, Pennsylvania, United States

About

I am a student studying AI and Robotics at the University of Pennsylvania, and a member of the Perception, Action, & Learning Group in Penn's GRASP Laboratory. This past year, I was a student researcher at the Allen Institute for AI on the PRIOR and Robotics teams, advised by Dr. Jiafei Duan, Prof. Ranjay Krishna, and Prof. Dieter Fox. There, I contributed to MolmoAct 2, a fully open robotics foundation model. Previously, in Summer 2025, I interned on Dyna Robotics' research team, where I trained frontier VLA models for real-world robotic deployment. At Penn, I have worked under the guidance of Dr. Jason Ma, Prof. Dinesh Jayaraman, and Prof. Osbert Bastani, alongside Will Liang and Johnny Wang. I've coauthored three papers on robot learning, which were published at ICLR, CoRL, and RSS. My earlier research on applying deep learning for rapid cardiac screening won a Second Grand Prize in Robotics and Intelligent Machines at ISEF 2023. Please feel free to contact me at [email protected]. Personal Website: https://samuelwang23.github.io/

Experience

  • Machine Learning Engineer Intern at Jane Street
    May 2026 - Present · 3 mos

  • Student Researcher at Ai2
    Sep 2025 - May 2026 · 9 mos

    Working on large-scale robotics foundation models. Advised by Dr. Jiafei Duan, Prof. Ranjay Krishna, and Prof. Dieter Fox.

  • University of Pennsylvania ()
    • Research Assistant in Perception, Action, & Learning (PAL) Group
      Oct 2023 - Present · 2 yrs 10 mos

    • Vice President of Engineering Dean's Advisory Board
      Sep 2023 - Present · 2 yrs 11 mos

    • Technology Chair of Penn Engineering Council
      Aug 2023 - Present · 3 yrs

  • Member of Technical Staff Intern at Dyna Robotics
    Jun 2025 - Aug 2025 · 3 mos

    Building general-purpose robots to power the future of the physical economy

  • Research Assistant in Vision, Imaging, and Data Analysis Research Lab at Temple University
    May 2023 - Aug 2023 · 4 mos

    - Developed computer vision algorithm for hotel chain recognition using PyTorch; Tracked experiments with Neptune - Created pipeline that extracts room objects with Segment Anything Model and DINO to improve model performance