Software Engineer - Reliability GPU Infrastructure

The Opportunity Luma AI is a capital-intensive lab building the future of creative intelligence. We have raised over $1.3 billion to fuel our mission. We operate with a lean, elite engineering team. This unique position offers you the leverage to build systems of immense scale while retaining individual ownership over the architecture and strategy of our infrastructure.

Where You Come In You will define the technical strategy for our compute substrate. This is a high-autonomy role where you will determine how we provision, manage, and scale our multi-cloud and on-premise GPU footprint. You will bridge the gap between hardware vendors and our software stack, ensuring our infrastructure evolves ahead of our model capabilities.

What You Will Build Hybrid Cloud Strategy: Architect a seamless infrastructure mesh that spans multiple cloud providers and bare-metal environments, optimizing for cost, performance, and reliability. Intelligent Scheduling: Design the logic that allocates massive compute resources across competing priorities, ensuring optimal throughput for both research training and production inference. Infrastructure as Software: Lead the effort to define our entire stack as code, building the rigorous CI/CD and GitOps workflows that allow us to move with speed and safety.

The Profile We Are Looking For Architectural Vision: You have a history of designing complex distributed systems, demonstrating the judgment to navigate trade-offs between immediate velocity and long-term scalability. Cloud Polyglot: You possess deep expertise across various infrastructure providers but understand the fundamental primitives well enough to build outside of them. Technical Leadership: You can mentor the team and drive consensus on technical decisions, setting the standard for engineering excellence in operations.