Software Engineer on Perception: machine learning and data infrastructure for perception foundational models

Machine Learning Software Engineer on Geo (Maps): applied computer vision and machine learning

Built machine learning systems to analyze StreetView imagery at a global scale

+ Update business information like hours and phone numbers to power autonomous map updates.

+ Find and understand all street signs in the world to power autonomous map updates.

https://blog.google/products/maps/how-ai-and-imagery-build-self-updating-map/

https://blog.google/products/maps/how-ai-and-imagery-keep-speed-limits-on-google-maps-updated/

+ "Learning the Complete Shape of Concentric Tube Robots" accepted by IEEE Transactions on Medical Robotics and Bionics (Volume: 2 , Issue: 2 , May 2020). Co-authored publication.

+ "Estimating the Complete Shape of Concentric Tube Robots via Learning" accepted by Hamlyn Symposium on Medical Robotics (HSMR) 2019. Co-authored publication.

+ Create a method for one robot to control another robot in order to implement Demonstrated Guided Motion Planning (DGMP). Acknowledged in many publications.

+ Programmed robots such as Baxter Robot and Fetch Robot using C++​ and Robot Operating System (ROS).

+ Created “ROBOPuppets” in order to control Baxter Robot using CAD, Arduinos​, and 3D printers.

Research and Development Intern on the Medical Team: Interactive Medical Simulation Toolkit (iMSTK).

www.imstk.org

+ Add a virtual peg transfer task trainer in iMSTK. The peg transfer task is a common training task for surgeons. (C++)

+ Use NVIDIA PhysX engine to simulate rigid body dynamics. (C++)

+ Provide force feedback through haptic devices such as the Phantom Omni.

+ Explore using AR hand tracking as an input to the simulation tool. (C++)

Software Engineering Intern on the Geo (Maps) Augmented Reality (AR) team.

+ Provide richer geotags for many of the 2.5T photos in Google Photos by integrating Mapping AR Localization techniques. (C++)