虚拟化研发工程师（GPU智算方向/基础技术） - AI算力基础设施
Core
Develop and optimize GPU virtualization technologies (vGPU, API forwarding) and resource pooling engines to enable fine-grained scheduling for AI training and inference, aiming for near-bare-metal performance.
Role type
Senior IC GPU virtualization engineer (AI infrastructure)
Builds
AI-native cloud infrastructure for LLMs, GPU resource pools, and high-performance computing platforms
Domain
Cloud infrastructure, AI computing, Operating Systems
Deliverable
production ML models | infrastructure
Required skills
C/C++ or Rust, Linux kernel internals, KVM/Virtio/VFIO, GPU virtualization (CUDA/GDS/GDR), PCIe/RDMA/CXL/NVLink protocols
Preferred skills
GPU performance tuning, DPU/smart NIC development, TensorFlow/PyTorch/vLLM/SGLang familiarity, Open source contributions (Linux kernel, QEMU, Rust-vmm)
Responsibilities
Implement and optimize GPU passthrough and vGPU solutions; Design GPU resource pooling components for memory slicing and compute splitting; Optimize data paths for AI frameworks to minimize virtualization overhead; Adapt and debug drivers for NVIDIA, AMD, and domestic GPUs; Troubleshoot Linux kernel, KVM/QEMU, and GPU driver issues
Seniority
Senior, hands-on IC
