Senior Staff Digital World System Engineer

XPENG is a leading smart technology company at the forefront of innovation, integrating advanced AI and autonomous driving technologies into its vehicles, including electric vehicles (EVs), electric vertical take-off and landing (eVTOL) aircraft, and robotics. With a strong focus on intelligent mobility, XPENG is dedicated to reshaping the future of transportation through cutting-edge R&D in AI, machine learning, and smart connectivity.

Job Responsibilities

• Be responsible for the overall architecture design and technological evolution of the next-generation synthetic data and Digital World generation system, supporting large-scale model training, simulation and evaluation in the directions of autonomous driving, robotics and embodied intelligence.

• Lead the construction of core capabilities in directions such as Digital City, Digital Twin, and World Model, and build a high-fidelity, highly scalable virtual world generation and closed-loop simulation system.

• Be responsible for the R&D of algorithms and systems related to 3D/4D scene reconstruction, including multi-view 3D reconstruction, Gaussian Splatting, NeRF, SLAM/SfM, spatiotemporal scene modeling and other directions, to build large-scale spatial understanding and reconstruction capabilities.

• Build a synthetic data production engine for perception, prediction, planning, VLA and robot model training, realizing high-quality, high-efficiency and low-cost data generation and delivery.

• Explore the application of cutting-edge technologies such as Generative AI, Diffusion, World Model and Physics Simulation in the field of simulation and data generation, and promote the evolution of data closed-loop towards automation and intelligence.

• Build a closed-loop simulation system for Digital City, realizing capabilities such as problem scenario reproduction, model playback verification, long-tail problem mining and automated evaluation.

• Collaborate in depth with teams of autonomous driving algorithms, robotics, simulation platforms, data Infra and training platforms to promote the integration and large-scale application of synthetic data and real data systems.

• Define synthetic data quality standards and evaluation systems, continuously optimize authenticity, diversity, generalization and Sim2Real effects, and improve the benefits of model training.

• Be responsible for the construction of synthetic data platform and team, promote the platformization, engineering and large-scale landing of core technologies, and form long-term evolution capabilities.

Requirements

• Master's degree or above in Computer Science, Artificial Intelligence, Robotics, Computer Vision, Computer Graphics or related majors.

• Have R&D experience in related fields such as synthetic data, simulation systems, 3D vision, neural rendering or embodied intelligence.

• In-depth understanding of technologies related to 3D reconstruction and spatial modeling, including but not limited to SfM, SLAM, NeRF, Gaussian Splatting, Differentiable Rendering, Photogrammetry and other directions.

• Candidates with relevant project experience in Digital Twin, Digital City, simulation platforms or large-scale virtual scene generation are preferred.

• Familiar with cutting-edge technology directions such as multi-modal large models, World Model, Generative AI, Diffusion and Reinforcement Learning, and have an in-depth understanding of data-driven model training systems.

• Have R&D experience in large-scale distributed systems, data generation platforms or high-pe