Tasks:
- * Build monitoring, alerting, log search, and fault diagnosis capabilities
- * Build operations agents to improve issue detection and resolution
- * Develop automated operations, inspection, and self-healing tools
- * Evolve cloud-native architecture for Kubernetes multi-cluster operations
- * Implement stability governance, incident response, capacity management, resource operations, and SLA management
- * Improve AI platform stability and availability
- * Operate large-scale GPU clusters, RDMA networks, and parallel storage
- * Troubleshoot GPU training, resource scheduling, network communication, and storage access failures
Perks/Benefits:
Skills/Tech stack required:
[AI Inference] [AI inference platforms] [AI Training] [AI training platforms] [Capacity Management] [Cluster operations] [Containerd] [Device plugin] [Distributed Training] [Golang] [GPU Cluster] [GPU Cluster Operations] [GPU communication] [GPU troubleshooting] [Grafana] [Hybrid Cloud] [Incident diagnosis] [Inference platforms] [Istio] [Karmada] [Kubernetes] [Linux] [Log search] [Loki] [MLOps] [Monitoring and Alerting] [NCCL] [NVIDIA device plugin] [Operations automation] [Parallel storage] [Prometheus] [Public Cloud] [PyTorch distributed] [PyTorch Distributed Training] [RDMA] [Resource scheduling] [Self-healing] [Self-healing systems] [Server Hardware] [Shell Scripting] [Training platforms]
Educational requirements:
N/A
Role(s):
[AI Platform Engineer] [DevOps Engineer] [Engineer] [Platform Engineer] [Reliability Engineer] [Site Reliability Engineer]