Senior OpenStack & GPU Infrastructure Architect | OCI Threads | Remote

OCI Threads; United States; Remote; USD 140,000 - 220,000; Full time

Job summary

OCI Threads is seeking a Senior OpenStack & GPU Infrastructure Architect to join our team remotely. You will own architecture workshops, create HLD/LLD, and drive IaC deployments across private cloud clusters featuring OpenStack and NVIDIA GPUs.

You will design compute, network, storage, and security architectures, perform capacity planning, validate driver stacks and DCGM, and lead complex troubleshooting in a fast-paced enterprise environment.

Qualifications

• Bachelor's degree in Computer Science, Electrical Engineering, Information Technology, or a related field.
• Minimum 8+ years of professional IT infrastructure architecture and cloud engineering experience.
• Strong, proven hands-on experience designing and implementing enterprise OpenStack private cloud environments.
• Deep technical knowledge of OpenStack compute, networking (Neutron), and storage architecture.
• Strong Linux infrastructure administration and automation / Infrastructure-as-Code (IaC) experience.
• Hands-on experience with NVIDIA GPU infrastructure, DCGM, Fabric Manager, and NCCL benchmarking.
• Demonstrated expertise in High Availability (HA), disaster recovery, performance tuning, and complex troubleshooting.
• Ability to independently develop comprehensive HLD, LLD, and implementation architecture documentation.
• Professional availability for remote employment.

Responsibilities

• Lead OpenStack architecture workshops, High-Level Design (HLD), and Low-Level Design (LLD) development.
• Design enterprise compute, network, storage, GPU acceleration, and security infrastructure architectures.
• Perform comprehensive capacity planning, infrastructure sizing, and hardware resource forecasting.
• Design and support hardened Linux-based OpenStack environments across control and compute planes.
• Lead Infrastructure-as-Code (IaC) based OpenStack deployment across control and compute nodes.
• Configure, administer, and troubleshoot OpenStack core services including Nova, Neutron, Cinder, Glance, Placement, and Horizon.
• Design and enable high-performance GPU workloads and container runtime integration.
• Review BIOS, BMC, GPU/HGX, NVMe, and NIC firmware compatibility across enterprise hardware.
• Implement, configure, and validate the NVIDIA driver stack, Fabric Manager, and Data Center GPU Manager (DCGM).
• Perform NCCL bandwidth testing, GPU burn-in, and performance validation across AI cluster nodes.
• Lead functional, performance, resiliency, failover, and disaster recovery testing.
• Provide L3 OpenStack architecture, platform troubleshooting, and operational support.

Skills

OpenStack

Neutron Networking

Linux Administration

Infrastructure-as-Code (IaC)

NVIDIA GPU infrastructure

DCGM

NCCL

HA / DR

Remote work

Documentation

Education

Bachelor's degree in CS/EE/IT

Tools

Fabric Manager

NVIDIA DCGM tooling

Kubernetes

Job description

Position Summary:

OCI Threads is seeking an experienced, highly technical, and accomplished Senior OpenStack & GPU Infrastructure Architect to join our team in a remote capacity. In this specialized private cloud and artificial intelligence infrastructure role, you will spearhead designing, deploying, validating, and optimizing large-scale OpenStack and NVIDIA GPU computing environments. You will work closely with cross-functional software architects, data science squads, and enterprise stakeholders to architect robust compute, network, storage, and GPU acceleration platforms. Ideal candidates bring a strong academic background in computer science or electrical engineering, deep practical mastery of hardened Linux systems, NVIDIA DCGM/NCCL toolchains, and OpenStack core services, with immediate availability for remote deployment.

Detailed Job Description:

As a Senior OpenStack & GPU Infrastructure Architect at OCI Threads in a remote capacity, you will take full ownership of leading architecture workshops, developing High-Level (HLD) and Low-Level (LLD) designs, and driving infrastructure-as-code deployments. Your day-to-day responsibilities encompass capacity planning, configuring OpenStack core components (Nova, Neutron, Cinder, Glance, Placement, Horizon), integrating NVIDIA GPU/HGX hardware, and validating driver stacks, Fabric Manager, and DCGM monitoring. You will perform NCCL bandwidth testing, GPU burn-in, resiliency validations, and complex troubleshooting across multi-node private cloud clusters. Working in a fast-paced enterprise environment, you will prepare comprehensive operational runbooks, conduct technical knowledge transfer sessions, and provide tier-3 architectural support.

Key Responsibilities:

• Lead OpenStack architecture workshops, High-Level Design (HLD), and Low-Level Design (LLD) development.
• Design enterprise compute, network, storage, GPU acceleration, and security infrastructure architectures.
• Perform comprehensive capacity planning, infrastructure sizing, and hardware resource forecasting.
• Design and support hardened Linux-based OpenStack environments across control and compute planes.
• Lead Infrastructure-as-Code (IaC) based OpenStack deployment across control and compute nodes.
• Configure, administer, and troubleshoot OpenStack core services including Nova, Neutron, Cinder, Glance, Placement, and Horizon.
• Design and enable high-performance GPU workloads and container runtime integration.
• Review BIOS, BMC, GPU/HGX, NVMe, and NIC firmware compatibility across enterprise hardware.
• Implement, configure, and validate the NVIDIA driver stack, Fabric Manager, and Data Center GPU Manager (DCGM).
• Perform NCCL bandwidth testing, GPU burn-in, and performance validation across AI cluster nodes.
• Lead functional, performance, resiliency, failover, and disaster recovery testing.
• Provide L3 OpenStack architecture, platform troubleshooting, and operational support.

Required Qualifications & Skills:

• Bachelor’s degree in Computer Science, Electrical Engineering, Information Technology, or a related field.
• Minimum 8+ years of professional IT infrastructure architecture and cloud engineering experience.
• Strong, proven hands-on experience designing and implementing enterprise OpenStack private cloud environments.
• Deep technical knowledge of OpenStack compute, networking (Neutron), and storage architecture.
• Strong Linux infrastructure administration and automation / Infrastructure-as-Code (IaC) experience.
• Hands-on experience with NVIDIA GPU infrastructure, DCGM, Fabric Manager, and NCCL benchmarking.
• Demonstrated expertise in High Availability (HA), disaster recovery, performance tuning, and complex troubleshooting.
• Ability to independently develop comprehensive HLD, LLD, and implementation architecture documentation.
• Professional availability for remote employment.

Nice-to-Have Skills:

• Professional certifications such as OpenStack Administrator (COA), Red Hat Certified Architect (RHCA), Linux Foundation Certified Engineer, or NVIDIA Deep Learning / Accelerated Computing credentials.
• Advanced familiarity with container orchestration platforms (Kubernetes) and cloud-native storage (Ceph).
• Experience integrating InfiniBand and high-speed RoCE (RDMA over Converged Ethernet) network fabrics.
• Prior working experience in artificial intelligence infrastructure startups, hyperscale cloud providers, or enterprise systems integrators.
• Contributions to open-source OpenStack projects or Linux kernel development.

Application Information:

• Recruiter: OCI Threads Recruitment Team
• Contact Name: Unspecified (OCI Threads Talent Acquisition)
• Email: [e-mail]
• Phone: Unspecified
• Application URL: Unspecified
• Salary/Rate: Competitive and commensurate with experience
• Deadline: Open until filled
• Notice Period: Immediate to short notice preferred
• Contract Duration: Permanent / Full-Time (Remote)

Recruitment Pro Tip:

When applying for senior OpenStack and GPU architecture roles, ensure your resume explicitly highlights your experience with NVIDIA DCGM/NCCL validation, automated IaC deployments, and multi-node private cloud sizing.