AI Infra SRE 工程师(Agent沙箱)
Core
Operate and maintain large-scale host clusters and control planes for AI agent sandboxes, ensuring high availability and stability.
Role type
Senior Infrastructure SRE (AI/Cloud Native)
Builds
Host clusters, container runtimes, and control plane components for AI agent sandboxes
Domain
Cloud Infrastructure / Virtualization / AI Systems
Deliverable
infrastructure
Required skills
Host cluster operations, Virtualization (KVM/QEMU/MicroVM), Container orchestration (K8s/Docker), System performance analysis (eBPF/perf), Go/Python/Shell scripting, Incident management
Preferred skills
LLM/AI Agent development, Capacity planning, Cost optimization
Technologies
KVM, QEMU, Firecracker, Kata, gVisor, Kubernetes, Docker, eBPF, perf, ftrace
Responsibilities
Maintain host clusters (kernel, virtualization, containers), define and monitor platform SLA/SLO metrics, manage incident/change/problem workflows, build observability and automation tools, implement AI-driven alerting and diagnostics, optimize capacity and costs
