AI Infra（云原生平台）研发实习生（J106487）
Core
Build and maintain a Kubernetes-based cloud-native platform for large model training and inference, managing container infrastructure, resource scheduling, and cluster operations.
Role type
AI Infrastructure (Cloud-Native Platform) Research Intern
Builds
Container infrastructure for AI workloads, cluster management tools, and resource scheduling systems
Domain
Cloud Computing / AI Infrastructure
Deliverable
infrastructure
Required skills
Kubernetes, Go, Linux, Container Orchestration, Distributed Systems, GPU Management, CNI/CSI, Operator Development
Preferred skills
Kubernetes Core Components, RDMA, Large Model Training/Inference, TP/DP/PP Parallelism, Open Source Contributions
Technologies
Kubernetes, Docker, containerd, Go, Python, C/C++, Linux, RDMA, GPU
Responsibilities
Develop container infrastructure for large model training and inference; Build cluster lifecycle management capabilities; Optimize core Kubernetes components (network, storage, scheduling); Manage heterogeneous compute resources (GPU, RDMA); Implement platform engineering for AI workloads; Develop observability and automated operations systems
