Vultr is hiring a
AI/ML Infrastructure Engineer

Summary

Join us in building and supporting our bare metal and GPU-based product offerings as a key member of the Engineering team. You will have ownership over the setup and provisioning of our GPU-based and bare metal systems, driving engineering and operational excellence around our infrastructure.

Requirements

• Hands-on experience working with current, high-performance GPUs, primarily NVIDIA products (e.g. NVLink, Infiniband, GRID drivers, vGPU and NVAIE)
• In-depth, hands-on experience working with and automating bare metal internals including BIOS, BMC, firmware, NICs, Redfish/IPMI, PCIe
• Experience with rail optimization across multiple clusters and architectures
• Experience with Linux, package management and device drivers
• Experience with commercial firmware
• Experience with Python, Bash, and PHP
• Experience with Machine Learning software

Responsibilities

• Developing and maintaining infrastructure in bare metal and containerized environments
• Work directly with our networking team to build scalable and supportable GPU clusters
• Ensure excellent customer experience by ensuring consistent and reliable provisioning of GPU infrastructure
• Build and maintain test automation of GPU-based products to ensure fast and reliable provisioning
• Implement and maintain GPU-based solutions to meet the needs of diverse applications and computational workloads
• Conduct in-depth benchmarking, performance testing, and troubleshooting of GPU systems to identify and resolve any hardware or software limitations
• Working with vendors to get all supported drivers and packages
• Working with vendors on any bugs, performance-related issues, hardware problems, and reference architectures
• Address any hardware, software, or performance issues promptly, coordinating with vendors, technical support, and internal teams as required

Benefits

$120,000 - $150,000