Inference Systems Engineer: GPU & Kubernetes Scale

Luma; United Kingdom; On-site; GBP 110,000 - 180,000; Full time

Job summary

Luma seeks a systems engineer to own large-scale model serving and inference pipelines. You’ll integrate new architectures into the inference engine, scale deployments across thousands of machines, and optimize GPU utilization while meeting SLOs.

You’ll collaborate across research, engineering, and infrastructure, build tooling to monitor lifetimes of inference jobs, and automate deployment pipelines for reliability and uptime in a multi-cluster, cloud- and on-prem environment.

Qualifications

• Strong Python and system-architecture skills.
• Experience deploying models with PyTorch, Hugging Face, vLLM, SGLang, or TensorRT-LLM.
• Experience with queues, scheduling, traffic control, and fleet management at scale.
• Experience with Linux, Docker, and Kubernetes, and with orchestration, deployment, and scheduling.
• Familiarity with Redis and S3-compatible storage.

Responsibilities

• Ship new model architectures by integrating them into the inference engine.
• Collaborate across research, engineering, and infrastructure to optimize model efficiency and deployments.
• Build internal tooling to measure, profile, and track the lifetime of inference jobs and workflows.
• Automate, test, and maintain inference services for maximum uptime and reliability.
• Manage and optimize inference workloads across clusters and hardware providers, and scale deployments across thousands of machines.

Skills

Python

System architecture

Linux

Scheduling at scale

Distributed systems

Docker

Kubernetes

Redis

S3-compatible storage

Tools

PyTorch

Hugging Face

vLLM

SGLang

TensorRT-LLM

Job description

Luma seeks a systems engineer to own large-scale model serving and inference pipelines. You’ll integrate new architectures into the inference engine, scale deployments across thousands of machines, and optimize GPU utilization while meeting SLOs.

You’ll collaborate across research, engineering, and infrastructure, build tooling to monitor lifetimes of inference jobs, and automate deployment pipelines for reliability and uptime in a multi-cluster, cloud- and on-prem environment.