大模型数据策略专家（VLA方向）-AI数据与安全
Core
Manage the full lifecycle of data collection, cleaning, annotation, and evaluation for VLA (Vision-Language-Action) models in embodied intelligence, supporting multimodal data including vision, proprioception, and motion.
Role type
Senior IC data strategy expert (VLA/Embodied AI)
Builds
High-quality multimodal datasets and data production pipelines for VLA model training
Domain
Embodied AI / Robotics / AI Safety
Deliverable
production ML models
Required skills
VLA model training methodology, multimodal data management, active learning, automated annotation, synthetic data generation, complex project management, cross-functional team leadership
Preferred skills
Experience with autonomous driving or embodied intelligence, VLM experience, technical insight into embodied AI research, product sense
Technologies
VLA models, VLM, active learning frameworks, simulation environments, annotation platforms
