AI GPU Infrastructure Operations Architect

Accrescent Group; San Francisco (CA); Presencial; USD 150 000 - 230 000; Tempo integral

Resumo da oferta

Accrescent Group in San Francisco seeks an experienced AI / GPU Infrastructure Operations Architect to deploy, troubleshoot, and optimize enterprise AI infrastructure. The role focuses on Linux-based infrastructure, Kubernetes and container platforms, networking, automation, monitoring, and operational support of AI/GPU environments.

The engineer will work closely with infrastructure, network, DevOps, cloud, security, and engineering teams to maintain platform availability, resolve complex

Qualificações

• Production Kubernetes experience including architecture, networking, storage, RBAC, pod lifecycle and cluster upgrades.
• Strong Linux administration and troubleshooting across compute, OS, and storage layers.
• Experience with AI/GPU platform environments and monitoring/observability tooling.

Responsabilidades

• Administer and support Linux and Kubernetes environments for AI/GPU workloads.
• Troubleshoot compute, OS, network, container, and platform issues; perform RCA.
• Support cluster deployment, upgrades, scaling, and maintenance activities.
• Automate health checks, monitoring, and remediation; enable CI/CD and IaC initiatives.
• Collaborate with cross-functional teams to maintain platform availability.

Conhecimentos

Linux administration

Kubernetes administration

Docker / container runtime

Networking (TCP/IP, DNS)

Shell / Bash scripting

Linux performance analysis

Incident management & RCA

Git / version control

Monitoring & troubleshooting

IaC tooling (Ansible / Terraform)

Ferramentas

Prometheus

Grafana

Splunk

Datadog

Dynatrace

Ansible

Terraform

AWS

Azure

GCP

Descrição da oferta de emprego

Job Description:

We are seeking an experienced AI / GPU Infrastructure Operations Architect with one of our client to support deployment, operations, troubleshooting, and reliability of enterprise AI infrastructure. The role will focus on Linux-based infrastructure, Kubernetes and container platforms, networking, automation, monitoring, and operational support of AI/GPU environments. The engineer will work closely with infrastructure, network, DevOps, cloud, security, and engineering teams to maintain platform availability, resolve complex infrastructure issues, perform root cause analysis, and improve operational reliability.

Key Responsibilities

AI / GPU Infrastructure Operations Support day-to-day operations of AI infrastructure and GPU-based platforms. Perform infrastructure health checks, validation, troubleshooting, and operational readiness activities. Troubleshoot infrastructure issues across compute, operating system, network, container, and platform layers. Support workload onboarding and platform configuration. Participate in incident resolution and infrastructure remediation. The existing AI Factory JD specifically identifies exposure to AI Infrastructure or GPU-based platforms as a preferred capability. Linux Administration Administer enterprise Linux environments including RHEL, Rocky Linux and Ubuntu. Troubleshoot OS, kernel, storage, networking, services, and application-related issues. Perform patching, upgrades, security hardening, and vulnerability remediation. Monitor system performance, capacity, and availability. Perform log analysis and troubleshoot recurring infrastructure issues. Manage system services, access, file systems, LVM, and packages. Kubernetes & Container Platform Administer and support Kubernetes environments. Support cluster deployment, upgrade, scaling, and maintenance activities. Troubleshoot Pods, Deployments, StatefulSets, Services, Ingress, Storage Classes, Persistent Volumes, and Namespaces. Investigate container runtime and orchestration issues. Maintain platform availability through proactive monitoring and health validation. Networking Troubleshoot infrastructure and Kubernetes network issues. Analyze TCP/IP connectivity, routing, DNS, load balancing, firewall, and ingress issues. Perform network validation and performance testing. Work with network teams on complex connectivity incidents. Perform packet-level troubleshooting when required. Internal AI Factory discussions also call out TCP/IP, VLANs, routing, link health, MTU, DNS/NTP, packet troubleshooting, fabric/GPU/host correlation, latency, throughput, packet loss, congestion and configuration drift. Automation & Scripting Develop Shell/Bash scripts for infrastructure and operational automation. Automate health checks, monitoring, reporting, log collection, and remediation activities. Support CI/CD and Infrastructure-as-Code initiatives. Identify opportunities to eliminate repetitive manual activities. Python scripting knowledge is desirable. Incident / Problem / Change Management Provide advanced technical support for infrastructure incidents. Resolve service disruptions within defined SLAs. Perform detailed Root Cause Analysis and implement preventive measures. Support production changes, validation, maintenance, and rollback activities. Act as an escalation point for operational issues. Monitoring & Reliability Support monitoring dashboards, alerts, and operational reporting. Analyze platform trends and identify reliability improvements. Reduce recurring incidents through automation and platform optimization. Support observability and proactive monitoring initiatives.

Skill Requirements:

Strong Linux administration: RHEL / Rocky Linux / Ubuntu Kubernetes administration and troubleshooting Docker / container runtime knowledge TCP/IP, DNS, routing, switching and load-balancing fundamentals Shell/Bash scripting Linux performance and log analysis Incident management and RCA Git/version control Infrastructure monitoring and troubleshooting The existing JD specifically calls for production Kubernetes experience including architecture, networking, storage, RBAC, pod lifecycle and cluster upgrades.

Other Requirements

AI Infrastructure / GPU platform exposure NVIDIA/GPU infrastructure concepts Prometheus / Grafana / Splunk / Datadog / Dynatrace Ansible / Terraform AWS / Azure / GCP CI/CD and DevOps practices Infrastructure-as-Code Experience working in large-scale production environments