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 Description Define the overall architecture of the vehicle-cloud integrated data closed-loop from a strategic perspective, making it the core infrastructure that supports weekly model iteration, cross-regional large-scale expansion, and safe, compliant, and auditable operations.

Lead the design of a highly available and scalable vehicle-cloud integrated data closed-loop system architecture, covering the entire link from on-vehicle data collection, encrypted upload, cloud access, preprocessing, storage, annotation scheduling, training data generation to simulation evaluation and feedback. At the same time, explore the next-generation AI Agent-centric data closed-loop architecture, formulate the architecture evolution roadmap from a strategic level, balance short-term delivery and long-term technical debt, and ensure the architecture has the scalability to meet business scale needs in the next 3-5 years. Focusing on the development direction of embodied intelligence and model self-training, continuously explore new AI technologies and data Infra toolchain development technologies to achieve efficient and high-quality flow of the entire link from data collection, data transmission, data processing, data clustering, data mining, data evaluation, data delivery to data effect feedback, and realize the continuous evolution of data lifecycle costs, data architecture and closed-loop engineering system.

Job Responsbilities Responsible for the design and optimization of the vehicle-cloud integrated data closed-loop architecture: Build and maintain the full-link large closed-loop system from on-vehicle data upload to cloud training and simulation evaluation, ensuring efficient and secure data flow between the vehicle and the cloud to support rapid model iteration. Build and maintain the data closed-loop toolchain: Lead the selection, development and integration of modules such as data processing links, data mining, collection and annotation tools, and visualization tools to improve the automation level and processing efficiency of data from original collection to usable data sets. Establish data lineage and version management mechanisms: Design and implement a data lineage tracking system to achieve full-process traceability of data from production, processing to use; establish strict corresponding relationships between data sets, annotation versions, and model versions to support problem attribution and iterative backtracking.

Explore the next-generation AI Agent-centric data closed-loop technology: Research and introduce AI Agent-based automated data processing and mining methods, explore the application of Agents in scenarios such as scene recognition, annotation assistance, and simulation use case generation, and promote the evolution of data closed-loop towards a higher level of intelligence. Support data work throughout the entire model development cycle: Deeply participate in the entire process of the model from data preparation, pre-training, fine-tuning, evaluation to on-board deployment and continuous optimization, understand the specific data needs of the model at each stage, and provide targeted data strategy support.

Senior Staff Physical AI Data Algorithm Engineer