AI Runtime Engineer
Core
Design and develop AI Runtime and inference infrastructure for autonomous driving machines and on-site edge computing platforms to enable high-performance, low-latency execution of ML, VLA, and LLM models.
Role type
Senior IC AI Runtime Engineer (Autonomous Driving & Edge AI)
Builds
Unified model loading, execution, version switching, resource management, health monitoring, and diagnostic capabilities for Perception, Prediction, VLA, and edge LLMs.
Domain
Autonomous driving, robotics, embedded systems, edge computing, high-performance computing
Deliverable
production ML models
Required skills
C++, Python, PyTorch model deployment, ONNX/TensorRT/TensorRT-LLM, GPU/NPU/SoC architecture, CUDA, model quantization, kernel optimization, Linux/container/embedded systems, system software concurrency and memory management
Preferred skills
Autonomous driving/robotics/industrial control edge AI deployment, NVIDIA Orin/Jetson/Drive/GPU Cluster experience, VLM/VLA/World Model/End-to-End autonomous driving/Local LLM inference, Functional Safety/Cybersecurity/Watchdog design
Technologies
PyTorch, ONNX, TensorRT, TensorRT-LLM, CUDA, Linux, Containers, C++, Python
Responsibilities
Design and develop unified model execution and resource management capabilities; Optimize end-to-end inference latency, throughput, memory, bandwidth, and power; Adapt runtime to various GPUs, NPUs, SoCs, and edge servers; Build runtime environments for VLA and edge LLMs; Establish runtime performance baselines and fault diagnosis mechanisms; Integrate models with on-board middleware, OS, and vehicle software for mass production.
Seniority
Senior, hands-on IC with technical leadership
