Inference Systems Engineer - Scale & GPU Fleet

Luma is seeking a systems engineer to own large-scale model serving, integrating new architectures into the inference engine, and scaling deployments across thousands of machines.

You will work with Kubernetes, Linux, Docker, and modern ML stacks to optimize throughput, reliability, and GPU utilization while meeting SLOs across diverse hardware providers, and contribute to the CI/CD of model checkpoints and SDKs.