Mid/Senior DevOps Engineer (Kubernetes, Linux)

Mô tả công việc

DevOps Engineer (Mid / Senior)

Datacenter Infrastructure & Observability

Location: Ho Chi Minh City, Vietnam (onsite)  |   Level: Mid / Senior   |   Type: Full-time   |   Experience: Mid 3+ years, Senior 5+ years

About SkyLab

SkyLab is a Singapore-headquartered AI infrastructure company operating across Southeast and North Asia. We design, build and operate GPU-as-a-Service (GPUaaS) environments on the latest NVIDIA platforms for AI model developers and enterprises, whose training and inference workloads depend on every node and link being healthy, 24 hours a day.

About the role

The DevOps team builds and runs the platform that keeps it that way. Our monitoring and automation stack watches hundreds of GPU servers across every site, covering hardware health, network fabric and alerting that gets the right people involved quickly. As new sites come online across the region, each one joins this platform from day one.

You will join the team that owns it end to end. Everything is infrastructure-as-code, reviewed through merge requests, and runs on-premises rather than on public cloud.

You will work closely with our Data Center Operations team and with engineers overseas, so clear written English is an important part of the job.

Key responsibilities

• Observability: operate and improve a multi-site monitoring platform, where each site keeps working on its own and a central NOC gives fleet-wide visibility.
• Hardware and network health: use out-of-band server data and network telemetry to spot issues with GPUs, nodes and links early.
• Automation: automate provisioning, configuration and deployment as code, and keep the CMDB accurate so it drives discovery and alerting.
• AI-augmented operations: apply AI tools and agents to reduce manual work, such as alert triage, incident summaries and runbook automation, and use AI assistants to write and review code faster.
• Alerting: tune alerts to be clear and actionable, and keep the runbooks behind them up to date.
• Troubleshooting: investigate issues across hardware, network, OS and application layers, find the root cause, and share what was learned.
• Reliability and security: help maintain highly available services, backup and recovery plans, secrets, and regular patching.
• CI/CD and tooling: improve pipelines and build internal tools that make changes easier and safer for other teams.
• Collaboration: document your work and partner with Data Center Operations and overseas engineers on design and problem-solving.
• Our toolset: Prometheus/Grafana stack, Terraform, Ansible, Kubernetes (k3s), Proxmox, GitLab CI, NetBox (CMDB), Jira and PagerDuty.