云原生 Agent Infra 高级研发工程师（J100117）
Core
Design and develop scalable cloud infrastructure for running AI Agents, including sandboxing, gateways, observability, and multi-agent orchestration.
Role type
Senior IC cloud-native infrastructure engineer (AI Agent platform)
Builds
Agent execution platform, sandbox environments, multi-agent orchestration systems, and developer tooling (CLI, SDKs)
Domain
Cloud-native infrastructure, AI Agent systems, distributed computing
Deliverable
production ML models | product features
Required skills
Go/Python/C++, Kubernetes, containerization (Docker/containerd), distributed systems, AI Agent concepts (MCP, Skill, context management), system design
Preferred skills
Virtualization/sandboxing, large-scale model training/inference optimization, serverless scheduling, observability (OpenTelemetry, Prometheus), full-stack development
Technologies
Kubernetes, Docker, containerd, KataContainer, gVisor, Firecriver, LangChain, LangGraph, AutoGen, Prometheus, Grafana, Loki, OpenTelemetry
Responsibilities
Design and develop Agent execution platform components (sandbox, gateway, observability); Implement high-security multi-tenant isolation and resource control mechanisms; Build multi-agent orchestration, scheduling, and governance capabilities; Construct developer interfaces (CLI, SDK, MCP) for Agent integration; Practice and promote Agentic workflows to solve production problems
Seniority
Senior, hands-on IC