腾讯云 EdgeOne-AI 推理基础设施高级工程师（上海/深圳）
Core
Design and develop the architecture for AI inference infrastructure and cluster management platforms to provide unified, stable, and scalable compute resources for various AI inference workloads.
Role type
Senior IC distributed systems engineer (GPU cluster management)
Builds
GPU resource management and scheduling systems, multi-cluster management platforms, and cluster lifecycle management tools.
Domain
Cloud infrastructure + AI inference
Deliverable
infrastructure
Required skills
Go, C++, Rust, Python, Kubernetes, YARN, Mesos, GPU scheduling, heterogeneous resource management, Gang Scheduling, GPU sharing, topology awareness, resource fragmentation governance, multi-cluster federation, capacity planning, fault migration, disaster recovery, multi-tenant governance, observability, automated recovery.
Preferred skills
Experience with complex production systems, cross-team project leadership.
Responsibilities
Design AI inference infrastructure architecture; build GPU resource management and scheduling capabilities; develop multi-cluster management and lifecycle tools; collaborate on multi-tenant governance and security; ensure platform stability and observability.
