About Luma AI Luma’s mission is to build multimodal AI to expand human imagination and capabilities. We believe that multimodality is critical for intelligence. This requires a massive, reliable, and performant GPU infrastructure that pushes the boundaries of scale. Our SRE team is the foundation of our research and product velocity, responsible for the thousands of NVIDIA and AMD GPUs across multiple providers that power our work.

Where You Come In We are looking for a hands-on, first-principles engineer who is fluent in Linux, comfortable operating close to the metal, and capable of architecting systems for the next generation of AI infrastructure.

You will build, maintain, and scale Luma’s infrastructure across on-prem and multi-vendor clouds (AWS & OCI), serving as the bridge between hardware vendors, cloud providers, and our research teams.

What You’ll Do

• Architect for Reliability & Scale: Participate in critical re-architecture sessions to redesign our systems for higher efficiency and scale. You won't just maintain existing clusters; you will help define how our next-generation infrastructure operates.
• Own Multi-Cloud GPU Clusters: Take end-to-end ownership of our production clusters for training and inference across AWS and OCI, ensuring high availability and peak performance.
• Drive Security & Compliance: Assist in achieving and maintaining security certifications (SOC 2 Type 1 & 2, ISO standards) by implementing robust infrastructure security practices in a fast-moving AI startup environment.
• Deep Linux Performance Tuning: Use your mastery of Linux systems to troubleshoot and optimize performance at the OS and kernel level.

Who You Are

• 8+ years of experience as an SRE, production engineer, or infrastructure engineer in a fast-paced, large-scale environment.
• Deep Linux Mastery: You possess deep, hands-on expertise in Linux, containerized systems, and debugging low-level system performance.
• Cloud Infrastructure Expert: You have strong experience with providers like AWS or OCI.
• Tenacious Troubleshooter: You thrive on solving complex, low-level problems where hardware and software intersect.
• Startup DNA: You are energetic and thrive in a less structured, fast-paced environment.
• Security-Minded: You possess a working knowledge of security best practices and familiarity with compliance frameworks, such as SOC 2 and ISO.
• Expert in High-Performance Networking: You have practical experience with InfiniBand, RDMA, or RoCE and understand how to optimize throughput for massive distributed training jobs.

What Sets You Apart (Bonus Points)

• Deep expertise with GPU tooling for NVIDIA and AMD GPUs like DCGM or ROCm.
• Experience managing large-scale GPU clusters for AI/ML workloads (training or inference).
• Familiarity with job management systems based on Kubernetes or orchestration frameworks like Ray.