Key skills
Artificial IntelligenceKubernetesGrafana
About this role
NVIDIA is leading the way in groundbreaking developments in Artificial Intelligence, High-Performance Computing and Visualization. They are seeking a highly motivated EngOps Engineer to join their advanced infrastructure software team, responsible for maintaining high-performance, rack-scale management solutions for datacenter environments.
Responsibilities:
- Take ownership of daily cluster failures and issues, troubleshooting them promptly to maintain optimal cluster availability and performance
- Manage updates to the site controller management nodes
- Manage the rollout and rollback of cluster software and firmware updates, ensuring smooth transitions and minimal disruptions
Requirements:
- BS or MS in Computer Science, Computer Engineering, Electrical Engineering, or a related field, or equivalent experience
- 5+ years of hands-on experience in deploying and administrating clusters, servers, switches, and related infrastructure
- Experience with deployment and configuration of operating systems, computer networks, and high-performance applications
- Proven ability to work effectively with developers and test engineers across different teams and time zones
- Experience deploying services in Kubernetes
- Datacenter or computer architecture experience is required—you should understand server, rack, and network topologies, as well as hardware/firmware/software interactions
- Background with hardware management protocols (Redfish, IPMI, BMC) and firmware update automation
- Experience configuring and debugging complex data center networks
- Experience developing scripts to automate recovery actions for management controllers and datacenter systems
- Direct experience with industry standard alerting tools and emergency response practices
- Experience with observability tools such as Grafana
- Hands-on experience with GPU-focused hardware and software, such as DGX systems and Compute Clusters
- Proficiency in designing large scale networking technologies and the associated challenges
- Experience with OpenStack and Foreman