Kumar Shankar
Senior Software Engineer at Overland AI
Redmond, Washington, United States
Summary
👤
Senior🎓
Top SchoolKumar Shankar is a Senior Software Engineer with 14 years of experience building robust, real-world autonomous robotic systems that tightly couple perception and planning. He holds a PhD in Robotics from Carnegie Mellon, where he developed quadrotor platforms, high-performance CUDA implementations for localization and mapping, and mentored junior researchers. At Microsoft he led and implemented end-to-end replay and KPI measurement frameworks for cross-platform spatial services and shipped practical fixes from kernel-level ARM32 bugs to Unity-based prototypes. Now at Overland AI he continues to bridge research and production, incorporating simulation-in-the-loop data generation and full-stack development across native Android and compliance domains. An active contributor to performance-focused open-source projects, he improved CUDA build tooling for NVIDIA's widely used gvdb-voxels repository, revealing his knack for low-level systems and reproducible builds.
14 years of coding experience
14 years of employment as a software developer
SanskritiB.E. Mechanical Engineering, B.E. Mechanical Engineering at Delhi College of EngineeringDoctor of Philosophy (Ph.D.) Robotics, Doctor of Philosophy (Ph.D.) Robotics at Carnegie Mellon UniversitySeth Anandram Jaipuria
Github Skills (8)
cuda10
cmake10
build-automation10
accelerated-computing9
gpgpu9
gpu9
c-language8
cprogramming-language8
Programming languages (5)
Github contributions (5)
NVIDIA/gvdb-voxels
Dec 2018 - Mar 2019
Sparse volume compute and rendering on NVIDIA GPUs
Role in this project:
Backend Developer
Backend DeveloperContributions:17 commits, 2 PRs, 30 comments in 3 months
Contributions summary:Kumar primarily focused on modifying and improving the build process for the NVIDIA GVDB-Voxels repository, specifically targeting CUDA-related build configurations. They implemented automatic CUDA architecture detection, added features for compiling PTX files, and fixed issues related to library dependencies. The user's contributions involved significant changes to CMake files and related build scripts, ensuring correct file generation and relative pathing.
cudanvidiagpusparsevolume
icoderaven/gvdb-voxels
Dec 2018 - Aug 2020
Sparse volume compute and rendering on NVIDIA GPUs
Contributions:46 pushes, 12 branches in 1 year 8 months
cudagpu-programmingnvidiagpusparse
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.