Kunimatsu Hashimoto
Physical AI & Robotics Engineer at Stealth-Mode Startup
Greater Boston United States
Summary
🤩
Rockstar🎓
Top SchoolKunimatsu Hashimoto is a Physical AI & Robotics Engineer with nine years of experience building sensor-driven robotic systems and model-based tools, currently working at a stealth-mode startup in Greater Boston. Previously a Staff Research Engineer at Toyota Research Institute after seven years as a senior researcher, he has deep expertise in integrating perception, rendering, and real-time camera pipelines for complex robotic platforms. He contributed notable backend work to the widely used Drake robotics library—implementing RGBD camera support, optimizing rendering and depth/label outputs, and adding tests and CMake integration—highlighting his focus on reliable sensor implementation. Trained in electrical and electronic engineering at Kyushu University, he combines hardware-aware systems thinking with software rigor to bridge simulation and physical robots. Colleagues rely on him for pragmatic solutions that improve performance and verifiability in perception stacks, and he often surfaces subtle trade-offs between sensor fidelity and runtime costs.
9 years of coding experience
9 years of employment as a software developer
Master's degree Electrical and Electronic Engineering, Master's degree Electrical and Electronic Engineering at Kyushu University
Github Skills (9)
robotics10
vtk10
c-language10
cprogramming-language10
image-processing10
testing9
cmake9
computer-vision8
opengl8
Programming languages (3)
Github contributions (5)
RobotLocomotion/drake
Jan 2017 - Dec 2021
Model-based design and verification for robotics.
Role in this project:
Backend Developer
Backend DeveloperContributions:18 reviews, 73 commits, 105 PRs in 4 years 11 months
Contributions summary:Kunimatsu primarily contributed to the development and implementation of the RGBD camera functionality within the Drake robotics library. Their work involved setting up the CMake build configuration for the RGBD camera, adding an image class, and creating unit tests. They also made substantial changes to the internal workings of the RGBD camera and related rendering system, including incorporating features like depth and label image outputs and camera pose updates, and also improving performance. This work demonstrates a focus on sensor implementation and rendering pipelines within the robotics context.
model-basedmodel-based-designroboticsdrakeverification
kunimatsu-tri/drake
Mar 2017 - Dec 2018
A planning, control, and analysis toolbox for nonlinear dynamical systems. More info at
Contributions:52 commits, 2 PRs, 409 pushes in 1 year 9 months
crosswordstructural-engineeringnonlineartoolboxdynamical-systems
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.