NOC Engineer, AI Infrastructure (L1 / L2 / L3)

About the Role

LanceSoft is building a NOC that monitors and supports AI and GPU-accelerated infrastructure for enterprise clients. This includes NVIDIA DGX and HGX systems, GPU clusters, high-speed networking, and AI workload platforms. You will be the front line for keeping these environments healthy, performant and available.

Required Certifications by Experience Level

• L1 Entry Level
  
• Required: NVIDIA-Certified Associate: AI Infrastructure & Operations (NCA-AIIO)
• Preferred: Candidates currently working toward the NVIDIA-Certified Professional: AI Operations (NCP-AIO) certification.

• L2 Intermediate Level
  
• Required: NVIDIA-Certified Associate: AI Infrastructure & Operations (NCA-AIIO)
• Preferred: NVIDIA-Certified Professional: AI Operations (NCP-AIO)

• L3 Senior Level
  
• Required: NVIDIA-Certified Associate: AI Infrastructure & Operations (NCA-AIIO) and NVIDIA-Certified Professional: AI Operations (NCP-AIO)
• Preferred: Additional certifications in cloud computing, networking, or related infrastructure technologies.

Common Responsibilities (All Levels)

• Monitor GPU clusters, servers, networks, storage and AI platforms through the NOC dashboards, and respond to alerts within SLA.
• Log, categorize, prioritize and track incidents in the ITSM tool (ServiceNow, Jira or similar).
• Follow runbooks and escalation matrices, and keep shift handover notes accurate.
• Communicate incident status clearly to internal teams and client stakeholders.
• Contribute to knowledge base articles and runbook improvements.
• Follow ITIL-aligned incident, problem and change practices.

L1 NOC Engineer: Monitoring and First Response

Experience: 0-2 years in NOC, IT support or data center operations

Responsibilities

• 24x7 monitoring of GPU health, utilization, temperature, power and node availability.
• First-level triage and validation of alerts. Perform basic checks such as node reachability, service status and job queue status.
• Execute documented remediation steps, such as restarting services or draining and returning nodes, following runbooks.
• Escalate unresolved or complex incidents to L2 with complete diagnostic details.
• Coordinate with data center or remote-hands teams on hardware tickets.

Skills

• Foundational understanding of AI infrastructure: GPUs, DGX and HGX systems, NVLink, InfiniBand and Ethernet fabrics.
• Basic Linux command line, networking fundamentals (TCP/IP, DNS, VLANs) and ticketing tools.
• Familiarity with nvidia-smi and basic monitoring dashboards.

L2 NOC Engineer: Investigation and Resolution

Experience: 3-5 years in infrastructure operations, with exposure to GPU or HPC environments

Responsibilities

• Investigate and resolve escalated incidents, including GPU faults, XID errors, driver and firmware issues, node failures and job scheduling problems.
• Administer and troubleshoot workload schedulers and orchestration (Slurm, Kubernetes with the NVIDIA GPU Operator).
• Analyze performance issues across compute, network (InfiniBand/RoCE) and storage, and perform root cause analysis.
• Manage cluster operations through NVIDIA Base Command Manager or equivalent tooling.
• Configure and tune monitoring and alerting (DCGM, Prometheus, Grafana) to reduce noise and improve detection.
• Execute changes, patching, firmware and driver updates under change control.
• Mentor L1 engineers and review the quality of escalations.

Skills

• Strong Linux administration and shell scripting (Bash, Python).
• Hands-on experience with GPU diagnostics, DCGM, container runtimes (Docker, containerd) and NVIDIA NGC containers.
• Working knowledge of InfiniBand and NCCL, and of storage for AI (parallel or NFS-based).

L3 NOC Engineer: Senior and Escalation Lead

Experience: 6-10 years in infrastructure or platform operations, including 2+ years on AI, GPU or HPC platforms

Responsibilities

• Act as the final technical escalation point for critical incidents and major outages. Lead incident bridges and post-incident reviews.
• Perform deep diagnostics on multi-node training and inference performance, fabric congestion, NVLink and InfiniBand faults, and GPU memory or ECC errors.
• Design and improve NOC monitoring architecture, observability, alert thresholds and SLO/SLA reporting for AI infrastructure.
• Automate operations using Python, Ansible and Terraform. Build self-healing workflows and reduce manual toil.
• Own problem management, capacity planning and lifecycle management (firmware, driver and CUDA compatibility matrices).
• Engage with NVIDIA and OEM support (Dell, HPE, Supermicro and others), and manage vendor cases through resolution.
• Define runbooks, standards and training plans for L1 and L2, and contribute to client solution reviews and pre-sales technical input where needed.

Skills

• Expert-level understanding of the NVIDIA AI stack: DGX, Base Command, DCGM, NCCL, GPU Operator, NGC, and NVIDIA AI Enterprise.
• Advanced Kubernetes and Slurm operations, and InfiniBand fabric management (UFM).
• Strong scripting and automation, with an observability and incident-command mindset.

Common Qualifications

• Bachelor's degree in Computer Science, IT, Electronics or a related field (or equivalent experience).
• The certifications listed for the level above (validity must be current).
• Willingness to work rotational shifts, weekends and on-call schedules.
• Strong written and verbal communication in English.

Nice to Have

• ITIL 4 Foundation
• Cloud certifications (AWS, Azure or GCP)
• CCNA or CCNP
• RHCSA or CKA
• Experience with air-gapped or on-prem AI environments
• Experience with the ServiceNow, Zabbix, SolarWinds or Nagios toolsets