TIONE-高级AI Infra后台研发工程师-K8S云原生方向-上海/杭州/深圳/北京
Core
Design and build a Kubernetes-based cloud-native inference platform for large language models, focusing on scheduling, resource management, and core plugin development to support high-elasticity, multi-tenant, and heterogeneous computing power.
Role type
Senior IC Kubernetes infrastructure engineer (LLM inference)
Builds
High-elasticity, multi-tenant LLM inference service platform
Domain
Cloud-native infrastructure, Large Language Models, Heterogeneous Computing
Deliverable
production ML models
Required skills
Go, Kubernetes (Scheduler Framework, CRD, Operator, Device Plugin, CSI, CNI), GPU/NPU resource management, High-performance networking (RDMA, RoCE), x86/GPU server architecture
Preferred skills
C++, Python, Overseas deployment experience, IDC construction support
Technologies
Kubernetes, controller-runtime, RDMA, RoCE, HGX, A100, H800, L20
Responsibilities
Lead K8s scheduling and resource management for heterogeneous compute; Design control and data plane architecture for the inference platform; Develop core CRDs and Controllers for lifecycle management; Build GPU/NPU device plugins and CSI/CNI plugins for storage and networking; Oversee overseas data center construction and GPU/network acceptance.
Seniority
Senior, hands-on IC
