GPU Devops Engineer

Role Overview

Seeking a skilled Software Engineer with expertise in GPU Infrastructure, Ubuntu operating systems, and automation to design, build, maintain, and optimize machine images across a variety of cloud platforms. In this role, you will play a key part in delivering standardized, secure, and high-performant images, including those deployed on GPU Bare Metal, GPU droplet (VM) images and CPU droplet images. Should also have working knowledge on linux, cloud, automation, image, GPU.

• Role: GPU Devops Engineer
• Location: All Persistent Locations
• Experience: 4 to 10 years
• Job Type: Full Time Employment

Responsibilities

• Work with key stakeholders in packaging and continuously testing our GPU images across multiple platform services.
• This includes coordination with hardware engineering teams on packaging firmware versions.
• Maintaining an active compatibility matrix of GPU drivers across Nvidia and AMD GPU platforms (H100/H200/MI300x/MI325x).

Requirements

• Strong experience with building Linux operating system images.
• Solid understanding of GPU related infrastructure including NCCL, RCCL testing frameworks particularly in a fabric connected environment.
• Hands-on experience building and managing machine images using tools like Packer, cloud-init, Ansible, Python, and shell scripting.
• Familiarity with CI/CD tools (e.g., GitHub Actions, Jenkins, GitLab CI) for automating image pipelines.
• Strong understanding of security practices in image building (e.g., CVE scanning, secrets management, system hardening).
• Automation and Scripting: Develop and maintain tooling and scripts (e.g., using HashiCorp Packer, Ansible, Terraform, Python, Shell scripting) to automate the creation, testing, and deployment of machine images.
• Security and Compliance: Ensure machine images adhere to security best practices, including hardening, patch management, and compliance with organizational and industry standards (e.g., CIS benchmarks, GDPR, HIPAA).
• Optimization: Optimize machine images for performance, size, and boot time to enhance scalability and reduce operational costs.
• CI/CD Integration: Integrate machine image creation into CI/CD pipelines using tools like Jenkins, GitHub Actions, or GitLab CI for automated builds and deployments.
• Versioning and Documentation: Maintain version control for machine images and document configurations, processes, and best practices.
• Monitoring and Troubleshooting: Implement monitoring for image performance and troubleshoot issues related to image deployment and runtime in cloud environments.
• Stay Updated: Keep abreast of advancements in cloud technologies, containerization, and image management tools to propose and implement improvements.