Dhiraj Maji
San Jose, California, United States
Summary
🎓
Top SchoolDhiraj Maji is a robotics and perception-focused software engineer with six years of experience building real-time vision systems for autonomous vehicles and industrial automation. He holds an MS in Robotics from the University of Michigan and has demonstrated production impact—migrating pointcloud pipelines to GPU (NVIDIA Orin/CUDA) to boost throughput by 54% and reduce CPU load, and shipping a YOLOv8-based pallet detector with MAP 0.91 for automated pick-and-place. His expertise spans LiDAR BEVs, multi-object tracking, sensor fusion (DETR + LiDAR), simulation-driven synthetic data pipelines in Isaac Sim, and low-latency system optimization. Currently at NVIDIA after leading perception at ArcBest, he combines academic research (event-camera stereo depth and multi-sensor calibration) with hands-on productionization of perception stacks. Notably, he has delivered practical tooling—interactive BEV visualizers and a 2-inch-accurate freight dimensioner—that directly improved teleoperation and planning in real deployments.
5 years of coding experience
2 years of employment as a software developer
Master of Science - MS, Robotics, 3.96/4.0, Master of Science - MS, Robotics, 3.96/4.0 at University of MichiganBachelor of Technology, Mechatronics Engineering, 9.29/10.0 (3.89/4.0), Bachelor of Technology, Mechatronics Engineering, 9.29/10.0 (3.89/4.0) at Manipal Institute of Technology
English, Bengali, HindiGithub contributions (5)
Contributions:12 commits, 9 pushes, 1 branch in 9 months
dhirajmaji7/QuickNAT-keras
Jun 2020 - Sep 2020
Contributions:10 pushes, 1 branch, 1 comment in 2 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.