Maggie Wang Email & Phone Number

Palo Alto, CA, US

San Mateo, California, US

• Worked on a diverse range of projects, including path planning, flight trajectory algorithms, and subject-tracking

• Designed and implemented robust deep-learning perception models for robot manipulation tasks by generating attacks in semantic space rather than pixel perturbations using the Drake dynamics and Blender 3D graphics.
• Searched for counterexamples that improve model performance using parallel black-box optimization (such as CMA-ES, RBFOpt, and Nelder-Mead) and adaptive-importance sampling methods
• Created keypoint detection pipeline of household objects using a Convolutional Neural Network
• Improved accuracy of pose estimation from keypoints using non-maximum suppression and least-squares optimization

San Mateo, California, US

• Helped unlock commercial and public safety opportunities by making robust docking a reality for Skydio 2, the most advanced autonomous drone in the market.

• Led a group of 15 students to compete in the AUVSI SUAS International Competition, placing 16th out of 75 teams in our first year.
• Created a fixed-wing plane capable of autonomous flight and navigation, remote sensing via onboard payload sensors (including a camera for vision processing with the NVIDIA Jetson), and a payload drop.

Mountain View, CA, US

• Implemented and tested embedded software (C++ and Python) for a new infrared sensor with 15x higher resolution for the Nest Learning Thermostat.
• Initiated and developed an R&D application in Python that uses vision processing and machine learning to streamline the thermostat setup process, with the potential of saving an estimated $1M a year for customer support.

Cambridge, MA, US

• Investigated autonomous underwater robots in order to better understand collective behavior in the Self-Organizing Systems Research Group.

Mountain View, California, US

• Led 85-member, all-girls robotics team / Girl Scout troop through design, build, and competition stages for the FIRST Robotics Competition.
• Used C++ to implement vision (with custom OpenCV code running on an NVIDIA Jetson TX1 coprocessor) and motion control algorithms on an ARM processor.
• Demoed robots at 30+ events, developed and taught 10+ programming workshops

Mountain View, California, US

Mountain View, California, US

• Started and led a team to compete in the Zero Robotics competition (http://zerorobotics.mit.edu/) - motivated by a problem of interest to DARPA, NASA and MIT.
• Formed alliances with teams from Germany, Italy, and US.
• Code ran on SPHERE satellites on the International Space Station.

Mountain View, California, US

PALO ALTO, California, US

• Award-winning martial artist for 12+ years, black belt since the age of 9.

Stanford, CA, US

• Converted control algorithm from MATLAB to C++ to run on Pixhawk PX4 quadcopters (drones).
• Wrote software for robust control of complex dynamical systems.

Santa Cruz, CA, US

• Studied the growth of galaxies at late stages using Python (2014).
• Studied the formation history of the Milky Way and Andromeda galaxies using Python and IDL (2015).
• Presented at 227th American Astronomical Society Conference and Bay Area Science Symposium (2016).
• Research through the Science Internship Program (SIP) - http://ucsc-sip.org/

• iOS Development Hackathon:- Award for Best New Developer (2011) for Decisioneer – helped users make decisions.- Award for Best Educational Application (2012) for TaekwondoKix – embedded videos, animations, and.
• Released Bubboo game on Google Play for Android (2014)

Davis, California, US

• Built robotic ping-pong shooter, simulated random walkers in C++.

Doctor Of Philosophy - Phd, Aeronautics And Astronautics

Physics A.B., Computer Science S.M.

Education record