Yunsuk Jeung
my picture
Yunsuk Jeung

Welcome to my blog!


Hello! I'm a software engineer passionate about designing and building robust and efficient systems. I enjoy digging into low-level performance tuning, memory efficiency, and system resilience to ensure software behaves predictably even under real-world constraints.

In my past roles, I honed my problem-solving skills by implementing technologies such as real-time SLAM, enabling systems to perceive and navigate complex environments.

To further develop my technical capabilities, I'm currently pursuing a Master’s degree at the University of Sydney.

Work Experience
MarsAuto
Software Engineer | Sep. 2024 - Feb. 2025
  • Engineered an automated pipeline generating high-fidelity 3D ground truth data, reducing data labeling time by 60% and accelerating
  • Implemented a deep learning-based matching framework and least squares refinement, improving 3D bounding box precision by 25% and reducing false positives by 18% in validation.
Maxst
Software Engineer | Feb. 2021 - Aug. 2024
  • Engineered high-performance SLAM and 3D reconstruction systems on smart glasses and mobile devices, achieving 90Hz visual-inertial SLAM with sub-centimeter accuracy and 30 FPS dense reconstructions with 4mm voxel resolution.
  • Improved localization for AR and autonomous vehicles by optimizing AR pipelines (reducing latency by 33ms) and fusing IMU–LiDAR–camera data, leading to 22% accuracy gain and 2× faster convergence
Publication
  • S. Choi, H. -W. Chae, Y. Jeung, S. Kim, K. Cho and T. -w. Kim, "Fast and Versatile Feature-Based LiDAR Odometry via Efficient Local Quadratic Surface Approximation," in IEEE Robotics and Automation Letters, vol. 8, no. 2, pp. 640-647, Feb. 2023