Xi Liu
PhD Candidate
Clemson, South Carolina, United States
Summary
🤩
Rockstar🎓
Top SchoolXi Liu is a PhD candidate at Clemson University specializing in generative AI, 3D generation and reconstruction, and 3D perception for autonomous driving, with five years of applied experience across industry and academia. He has contributed to production-grade perception stacks—working as a LiDAR perception engineer at Momenta and as an applied scientist intern on Amazon’s Just Walk Out team—bridging research prototypes and deployable systems. Xi has a strong open-source footprint in OpenMMLab’s mmdetection3d, where he improved CUDA implementations, unit tests, and dataset processing to boost reliability and GPU compatibility. His research roles include dynamic 3D scene reconstruction at the Pioneer Centre for AI and 3D object detection work at CUHK, reflecting a focus on both fundamental algorithms and system-level robustness. Educated in China and Denmark, he combines international academic training with hands-on debugging and optimization skills that often uncover subtle data-format and GPU edge cases.
5 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD, Doctor of Philosophy - PhD at Clemson UniversityMaster of Science - MS Computer Science, Master of Science - MS Computer Science at Københavns Universitet - University of CopenhagenComputer Science and Technology, Computer Science and Technology at Jilin University
Github Skills (12)
cuda10
unit-testing10
object-detection10
pytorch10
test-integration10
3d-object-detection10
point-cloud10
test-unit10
python10
integration-testing10
testing10
numpy9
Programming languages (4)
Github contributions (5)
open-mmlab/mmdetection3d
Jan 2021 - Aug 2021
OpenMMLab's next-generation platform for general 3D object detection.
Role in this project:
ML Engineer & QA Engineer/Test Automation Engineer
ML Engineer & QA Engineer/Test Automation EngineerContributions:49 reviews, 30 commits, 60 PRs in 7 months
Contributions summary:Xi primarily contributed to unit testing and debugging core functionalities within the 3D object detection framework. They implemented and refined unit tests for voxelization, and dataset processing, ensuring the correctness of critical operations. Furthermore, the user addressed bugs and optimized CUDA implementations, indicating a focus on improving the framework's performance and reliability. They also added tests to ensure compatibility with GPU-based computations, and fixed issues related to data format in various testing scenarios.
3d-object-detectionopenmmlabpytorchobject-detectionpoint-cloud
xiliu8006/mmdetection3d
Jan 2021 - Feb 2022
OpenMMLab's next-generation platform for general 3D object detection.
Contributions:441 commits, 3 PRs, 358 pushes in 1 year 1 month
3d-object-detectionopenmmlab
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.