Edge Computing Researcher (Physical AI & Robotics)

About the RoleWe are seeking passionate and innovative researchers to advance Edge Computing technologies for Physical AI and robotics applications.Our research focuses on bringing advanced AI capabilities closer to the physical world, enabling robots and intelligent machines to perceive, reason, and act efficiently through optimized algorithms, accelerated inference, and heterogeneous computing architectures.Key Responsibilities1. Edge AI Computing & Heterogeneous ArchitectureResearch efficient execution of advanced AI models at the edge, including multimodal foundation models, Vision-Language-Action (VLA) models, and World Action Models (WAMs) etc.Research model partitioning, distributed inference, and hardware-aware optimization across edge devices and infrastructure.Investigate inference acceleration and heterogeneous computing architectures, including CPUs, GPUs, NPUs, FPGAs, and emerging specialized accelerators, balancing performance, latency, energy efficiency, and resource constraints.2. Edge-Enabled Physical AI SolutionsDevelop edge computing solutions for multimodal sensor fusion, environmental perception, intelligent decision-making, and robotic manipulation.Support emerging applications involving embodied intelligence, dexterous manipulation, and intelligent robotic systems.3. Research & PrototypingIdentify research opportunities, propose innovative solutions, and develop proof-of-concept prototypes.Evaluate performance, efficiency, and practical feasibility, contributing to publications, patents, and technology demonstrations.QualificationsEducation: PhD in Computer Science, Computer Engineering, Electrical Engineering, Artificial Intelligence, Robotics, or related disciplines preferred. Candidates with a Master's degree and outstanding technical or research achievements are also welcome.Core Technical Expertise: Research or engineering experience in at least one of the following Edge Computing areas: AI Inference Optimization: Model partitioning, distributed inference, model compression, or efficient execution of AI workloads.Heterogeneous Computing & Acceleration: Hardware-software co-design, workload mapping, and acceleration using CPUs, GPUs, NPUs, FPGAs, or other computing architectures.Edge & Distributed AI Systems: Edge computing architectures, computation offloading, resource scheduling, or optimization of latency, throughput, memory, and energy efficiency.Physical AI Applications: Interest in applying edge computing technologies to robotics, embodied AI, multimodal intelligence, or real-time intelligent systems. Prior experience in these applications is an advantage.Research & Problem-Solving: Strong analytical and innovative thinking, with the ability to address challenging technical problems. Impactful research contributions, especially highly cited publications, are highly valued.Technical Implementation: Practical experience with relevant programming languages, AI frameworks, hardware platforms, or development toolchains. Strong hands-on experience in building, optimizing, or demonstrating innovative AI or computing systems is highly valued.