AI Infrastructure Engineer
Core
Design, implement, and optimize high-performance, scalable inference systems for Large Language Models (LLMs) and Vision-Language Models (VLMs) across cloud, edge, and hybrid environments to power intelligent automotive systems.
Role type
Senior AI Inference Infrastructure Software Engineer
Builds
Production-grade AIOS platform inference kernels and serving systems for NIO's smart electric vehicles
Domain
Automotive AI / High-performance Systems Engineering
Deliverable
production ML models
Required skills
C/C++, CUDA, PyTorch, GPU/NPU programming, kernel development, distributed inference, memory optimization, computer architecture, Transformer internals
Preferred skills
Hardware-aware model optimization, edge/embedded AI, inference serving systems (batching/scheduling), open source contributions
Technologies
CUDA, PyTorch, TensorFlow, GPU, NPU, DSP
Responsibilities
Design scalable inference systems for LLMs/VLMs; Develop custom kernels for hardware accelerators; Integrate optimization techniques (KV-cache, tensor parallelism, quantization); Partner with system/hardware teams for integration; Translate requirements into production-ready software; Drive evolution roadmap for AIOS stack
Seniority
Senior, hands-on IC