Maggie Wang

robotics, space exploration, photography, writing

I have always been drawn to building and creating—starting with arts and crafts, then scientific experiments, and eventually complex engineering systems. I am an artist and writer at heart, a scientist and engineer by training. Above all, I am driven by curiosity and exploration.

I am currently working on making vision-language-action models more steerable to help with skill acquisition. Feel free to reach out if you’d like to connect about research or potential collaborations! 😊

🤖 PhD Projects

• Phys2Real: Fusing VLM Priors with Interactive Online Adaptation for Uncertainty-Aware Sim-to-Real Manipulation — project website, paper (Oct 2025, ICRA 2026)
  • Early workshop version: Physically-Informed Gaussian Splatting for Adaptive Sim-to-Real Transfer in Robotic Manipulation, RSS Structured World Models for Robotic Manipulation (June 2025)
• NASA NSTGRO24 Fellow, in collaboration with NASA JPL — Adaptive World Models for Space Robotics: Implicit Representations Grounded in Semantics and Physics
• Workshop organizing:
  • Generative Digital Twins for Real2Sim and Sim2Real Transfer in Robotics at ICRA 2026 — website (June 2026)
  • 2nd Space Robotics Workshop (SRW), Session Chair for Unlocking the Offworld Economy with Robotic Technologies at IEEE SMC-IT/SCC — recordings and details (July 2025)

✈️ Industry Experience

• Designed and implemented a global graph-based planner, improved core flight algorithms for precise movement and multi-frame tracking, and trained optical flow models for subject localization at Skydio (Fall 2021-Spring 2023). Talk at the Women in Data Science Conference (2022).
• Docking system for autonomous drones at Skydio (Summer 2019)
• Embedded systems and vision processing for smart homes at Nest/Google (Summer 2018)

🎓 Undergraduate Research & Projects

• Robust pose estimation in perception systems (keypoint detection) for robot manipulation at the MIT Robot Locomotion Group (paper, slides) (Fall 2019 - Spring 2021)
• Illegal gold mining barge detection in the Rio Madeira using data from Sentinel-1 (radar) and Planet Labs (optical) using both traditional computer vision and deep learning techniques (paper) (Spring 2021)
• Led team of ~15 undergraduates that built fixed-wing planes in the Harvard Undergraduate Robotics Club’s Unmanned Aerial Systems team (now the Harvard Undergraduate Aeronautics Club) (Fall 2017 - Summer 2019) – flight readiness video and promotional video for the AUVSI Competition in 2019
• Helped start the Harvard Satellite Team (Fall 2018)

🚀 Early Projects

• Control algorithms for drones at the Stanford Autonomous Systems Laboratory (Summer 2016)
• Programming Captain - built robots with FIRST Robotics Team 1868, The Space Cookies (Summer 2013 – Spring 2017)
• Project Leader for MIT Zero Robotics - algorithms ran on SPHERE satellites on the International Space Station (Summer 2014 - Winter 2015)
• Galaxy formation and evolution using data from the Hubble Space Telescope in the Department of Astrophysics at UC Santa Cruz through the Science Internship Program (Summers 2015, 2016) – Studying the Structure and Dynamics of the Subcomponents of the Andromeda Galaxy