Devops Engineer (GPU Bare Metal)

Cloud Hybrid Technologies, LLC; Houston (TX); On-site; Full time

Job description

We need engineers skilled in multiple areas here to support both GPU Bare Metal and GPU VM products. We also need engineers skilled in AI/ML DevOps and Linux Sysadmins.

GPU Bare Metal

Required Skills

• Proven ability to orchestrate bare metal linux systems at scale including building automation for firmware updates, bios config management, configuring PXE environments.
• Deep Linux systems experience including troubleshooting network interfaces, developing and applying configuration management, security best practices and monitoring and alerting.
• Strong automation mindset. Expert knowledge in 1 or more orchestration tools such as MaaS, Salt, Chef, Ansible or Puppet.
• Strong communication skills. Your job will involve writing detailed documentation for others to pick up or leading knowledge sharing sessions with operations teams.
• Hands-on experience in High Performance Computing (HPC) clustered environments from Nvidia or AMD. Experience in performing automated wide scale testing on NCCL or other frameworks.
• Network engineering experience with VyOS platforms.

What You\'ll Be Working On:

• Provisioning and automating GPU Bare Metal deployments
• DevOps - Assist customer support and CloudOps teams with GPU specific knowledge/debugging during customer escalations
• Performance testing, analysis and monitoring
• Firmware, BIOS, Kernel upgrades and testing

GPU Virtual Machines

Required Skills

• Strong understanding of Linux based operating systems
• Deep experience with the internals of QEMU, KVM, Linux kernel and libvirt. Strong proficiency in C.
• Strong knowledge of DO’s proprietary services and how they intersect with our virtualization stack.

Cloud Hybrid is an equal opportunity employer inclusive of female, minority, disability and veterans, (M/F/D/V). Hiring, promotion, transfer, compensation, benefits, discipline, termination and all other employment decisions are made without regard to race, color, religion, sex, sexual orientation, gender identity, age, disability, national origin, citizenship/immigration status, veteran status or any other protected status. Cloud Hybrid will not make any posting or employment decision that does not comply with applicable laws relating to labor and employment, equal opportunity, employment eligibility requirements or related matters. Nor will Cloud Hybrid require in a posting or otherwise U.S. citizenship or lawful permanent residency in the U.S. as a condition of employment except as necessary to comply with law, regulation, executive order, or federal, state, or local government contract