DevOps 工程师（AI 推理）
DevOps Engineer (AI Inference)
作为DevOps工程师，你将负责设计、部署和维护在本地环境中支持可扩展且安全的AI推理工作负载的基础设施和服务。
你将负责的工作
- 设计、开发和维护AI推理工作负载的基础设施，包括GPU调度、模型部署流水线以及本地环境中的数据访问模式
- 构建和管理AI推理平台的监控和可观测性工具，包括用于模型健康状况和系统性能的仪表板、警报和操作手册
- 与机器学习工程师和平台团队合作，设计AI工作负载的系统架构，集成推理运行时，并进行大规模性能测试
我们寻找的人选
- 对Kubernetes架构有深入理解，包括CNI、CSI、Operators、入口/网关以及控制平面组件
- 有操作和排查生产环境Kubernetes集群的实际经验
- 强大的Linux和网络故障排除技能，包括DNS、路由、防火墙、TLS、MTU、连接性和性能问题
- 能够使用Python、Go或Bash开发自动化和运维工具
- 有使用Terraform、Ansible或类似IaC/配置管理工具的经验
- 有使用VictoriaMetrics/Grafana或类似监控、告警和故障排除工具的经验
- 熟悉基于Git的工作流程和CI/CD流水线
优先考虑的资质
- 熟悉Cluster API或类似的Kubernetes集群生命周期管理技术
- 有操作或管理Slurm集群的实际经验
- 了解Argo CD、GitOps工作流、Helm或Helmfile
- 有在托管平台、PaaS或云服务中工作的背景
- 接触过裸金属、GPU、HPC或其他高性能计算环境
加分项
- 熟悉NVIDIA GPU堆栈、RDMA/InfiniBand或高性能网络
- 了解OpenStack或类似的云基础设施平台
- 有开发Kubernetes Operators或Controllers的实际经验
福利待遇
在Gcore，我们希望你能发挥最佳水平并享受这段旅程。我们的福利旨在支持你的成长、健康和工作之外的生活：
- 具有竞争力的薪酬
- 根据职位灵活的工作时间和混合或远程办公选项
- 每年最多可在世界任何地方工作45天
- 为本人及家人提供的私人医疗保险*
- 额外的带薪年假和病假天数
查看英文原文
As a DevOps Engineer, you will be responsible for designing, deploying, and maintaining infrastructure and services that enable scalable and secure AI inference workloads on-premises.
What You Will Do
- Design, develop, and maintain infrastructure for AI inference workloads, including GPU scheduling, model deployment pipelines, and data access patterns in on-prem environments
- Build and manage monitoring and observability tools for AI inference platforms, including dashboards, alerts, and runbooks for model health and system performance
- Collaborate with ML engineers and platform teams to design system architecture for AI workloads, integrate inference runtimes, and test performance at scale
What We're Looking For
- Strong understanding of Kubernetes architecture, including CNI, CSI, operators, ingress/gateway, and control plane components.
- Hands-on experience operating and troubleshooting production Kubernetes clusters.
- Strong Linux and networking troubleshooting skills, including DNS, routing, firewalling, TLS, MTU, connectivity and performance issues.
- Ability to develop automation and operational tooling using Python, Go, or Bash.
- Experience with Terraform, Ansible, or similar IaC/configuration management tools.
- Experience with VictoriaMetrics/Grafana or similar monitoring, alerting, and troubleshooting tools.
- Strong experience with Git-based workflows and CI/CD pipelines.
Preferred Qualifications
- Familiarity with Cluster API or similar Kubernetes cluster lifecycle management technologies.
- Hands-on operation or administration of Slurm clusters.
- Knowledge of Argo CD, GitOps workflows, Helm, or Helmfile.
- Background working with managed platforms, PaaS, or cloud services.
- Exposure to bare metal, GPU, HPC, or other high-performance computing environments.
Nice to Have
- Familiarity with the NVIDIA GPU stack, RDMA/InfiniBand, or high-performance networking.
- Knowledge of OpenStack or similar cloud infrastructure platforms.
- Hands-on experience developing Kubernetes operators or controllers.
Benefits
At Gcore, we want you to do your best work and enjoy the journey. Our benefits are designed to support your growth, well-being, and life beyond work:
- Competitive compensation
- Flexible working hours and hybrid or remote options, depending on your role
- Work from anywhere in the world for up to 45 days per year
- Private medical insurance for you and your family*
- Extra paid vacation and sick leave days*
- Support for life’s important moments and celebrations
- Language courses to help you connect and grow
- Modern, welcoming offices with snacks, drinks, and entertainment*
- Team sports and social activities*
*Benefits may vary depending on your location.
Equal Opportunity Employer
We provide equal opportunity to all applicants without regard to race, color, religion, sex, sexual orientation, age, gender identity, gender expression, national origin, disability, or any other legally protected characteristics.
This position is available only under an employment (labor) agreement.
The world’s digital experiences run on something invisible: the infrastructure and software that keep them fast, reliable, and secure. At Gcore, you’ll help design and deliver that foundation for an AI-driven world.
We’re a global provider of infrastructure and software solutions for AI, cloud, network, and security, powering everything from real-time communication and streaming to enterprise AI and secure web applications. With 210+ edge locations, 50+ cloud regions, and thousands of GPUs, your work here can reach users and businesses across the globe.
You’ll collaborate with leading technology partners such as Intel, NVIDIA, Dell, and Equinix, and work on platforms that power digital products used around the world. Our vision is simple: to connect the world to AI, anywhere, anytime.
Want to work on technology that goes beyond a single product or industry? Join a global team of 550+ professionals building infrastructure and software that supports the entire digital ecosystem.
We are looking for a talented DevOps Engineer to join our AI Inference Operations Team.
Originally posted on Himalayas