Yukihiro Saito
Architect at TIER IV
Chiyoda, Japan
Summary
🤩
Rockstar🎓
Top SchoolYukihiro Saito is an architect with 11 years of experience specializing in autonomous driving and robotics, combining deep learning, ROS1/ROS2, and real-time embedded systems to deliver production-ready perception, planning, and localization stacks. At TIER IV he progressed from engineer to architect, leading teams and contributing substantial back-end improvements to the widely used Autoware ecosystem—enhancing shape estimation, camera-LiDAR fusion, tracking gates, and EKF tuning. His background spans R&D and leadership roles (including a CTO stint) and a master’s in information science, giving him both hands-on algorithmic depth and product-focused system design skills. Known for pragmatic debugging tools and single-frame occupancy processing optimizations, he tends to bridge research ideas and robust, real-time implementations.
11 years of coding experience
2 years of employment as a software developer
Master's degree, Information science and engeering, Master's degree, Information science and engeering at 立命館大学
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (12)
object-tracking10
c-language10
cprogramming-language10
ros10
autonomous-driving10
algorithms9
kalman-filter9
point-cloud9
data-structure9
algorithm9
data-structures9
computer-vision9
Programming languages (13)
Github contributions (5)
autowarefoundation/autoware_universe
Feb 2022 - Jan 2023
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1435 reviews, 36 commits, 145 PRs in 10 months
Contributions summary:Yukihiro primarily contributed to the `autoware_universe` repository by modifying and improving the shape estimation and detection by tracker modules. Their work involved refining algorithms for object shape and pose estimation, particularly for vehicle types, along with adding debugging tools and refining the intersection over union (IoU) calculations. They also worked on integrating camera-lidar fusion for improved object recognition. Further work included adding support for single-frame processing in the occupancy grid, adding an IoU gate for tracking, object validation, and updating EKF parameters.
autowareros23d-mapcalibrationros
tier4/sensor_component_description
Sep 2021 - Apr 2024
Contributions:2 reviews, 4 PRs, 4 pushes in 2 years 6 months
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.