Simon Stepputtis

About

I am currently a Project Scientist at Carnegie Mellon’s Robotics Institute, but I will start as Assistant Professor at Virginia Tech in Fall 2025. In my research, I aim to create intelligent robots and AI systems that are capable of seamlessly integrating with human environments and tasks. To this end, I develop neuro-symbolic methods that reason over the attributes and affordances of objects in the environment, allowing them to learn behaviors efficiently and adapt to new environments. I earned my Ph.D. from Arizona State University under Prof. Heni Ben Amor in the Interactive Robotics Lab, where I specialized in physical human-robot interaction. I’ve also collaborated on projects with Bosch, Intel AI Labs, and X, The Moonshot Factory. My work has been featured at top venues such as NeurIPS and Autonomous Robots, earning spotlight presentations and the Best Poster Award at the Southwest Robotics Symposium.

Experience

  • Assistant Professor (Starting in Fall 2025)

    • Virginia Tech
    • I am starting a new research lab for intelligent robotics. Stay tuned for more!

  • Postdoctoral Fellow / Project Scientist (2022 - 2025)

  • Resident @ Google X (May to October 2021)

    • X, The Moonshot Factory
    • As a resident, I worked on industrial manipulation tasks for Intrinsic, a robotics software and AI project at Google X.

  • Robotics Intern (Summer 2018)

    • Robert Bosch LLC
    • During my interhsip, I segmented time series data into semantic sections and validated user behaviors based on usage patterns and global goal constraints.

Education

  • Ph.D. Computer Science (2017 - 2021)

    • Arizona State University, USA
    • Thesis: Multimodal Robot Learning for Grasping and Manipulation

  • M.Sc. Engineering & Computing (2015 - 2016)

    • Technische Universität Freiberg, Germany
    • Thesis: A data-driven approach for triadic interactions in human-robot interaction

  • B.Sc. Engineering & Computing (2011 - 2015)

    • Technische Universität Freiberg, Germany
    • Thesis: Upper body tracking for avatar visualization in HMD-based virtual reality