Latest AI JobsBrowse Jobs
Luma AI logo

Software Engineer, Inference

Luma AILondon, UKResearch & AI
Apply on company site
Salary
Not listed
Job type
Full-time
Work mode
Remote
Experience
Senior
Posted
8 Oct 2026
Status
Open · checked 7d ago

About this role

This systems engineering position is suited for a strong systems engineer comfortable with model serving and Kubernetes at scale who wants to focus on the infrastructure side rather than pure modeling. Located in London, United Kingdom, this full-time role at Luma AI involves owning the serving of generative AI models, scaling deployments, and maximizing GPU fleet utilization.

Role Overview

Luma AI is hiring a Software Engineer for Inference in London, UK. The position centers on large-scale inference systems work, covering scheduling, fleet management, deployment pipelines, and operational reliability across clusters and hardware providers.

Key Responsibilities

Deliver new model architectures by integrating them directly into the inference engine.
Work alongside research, engineering, and infrastructure teams to enhance model efficiency and deployments.
Construct internal tooling to measure, profile, and monitor the lifetime of inference jobs and workflows.
Automate, test, and maintain inference services to ensure maximum uptime and reliability.
Oversee and optimize inference workloads across diverse clusters and hardware providers while scaling deployments across thousands of machines.
Design scheduling systems that utilize expensive GPU resources efficiently while respecting internal SLOs, alongside maintaining CI/CD for model checkpoints and SDKs.

Requirements

Strong background in Python and system-architecture.
Practical experience deploying machine learning models using tools like PyTorch, Hugging Face, vLLM, SGLang, TensorRT-LLM, or comparable technologies.
Background working with queues, traffic control, scheduling, and fleet management at scale.
Proficiency with Linux, Docker, Kubernetes, and associated orchestration, deployment, and scheduling tools.
Familiarity with Redis and S3-compatible storage solutions.

Nice to Have

Experience with modern networking stacks such as RDMA, including RoCE, InfiniBand, and NVLink.
Familiarity with high-performance, large-scale machine learning systems managing over 100 GPUs.
Working knowledge of CUDA, FFmpeg, or multimedia processing.

Skills and Tools

Python
PyTorch
Hugging Face
vLLM
SGLang
TensorRT-LLM
Linux
Docker
Kubernetes
Redis
S3-compatible storage
RDMA, RoCE, InfiniBand, NVLink, CUDA, FFmpeg

About the Company

Luma AI aims to build unified general intelligence capable of generating, understanding, and operating within the physical world. The organization focuses on multimodality as a crucial step beyond language models by incorporating vision. Luma AI operates as an equal opportunity employer.

How to Apply

Interested candidates should apply directly on the official Luma AI careers page.

Skills and tech stack

PythonPyTorchHugging FacevLLMSGLangTensorRT-LLMLinuxDockerKubernetesRedisS3RDMARoCEInfiniBandNVLinkCUDAFFmpegLLM

Questions candidates ask

Where is this job located?

This role is located in London, UK.

Does this role focus on machine learning modeling or infrastructure systems?

This position focuses firmly on the systems side, covering scheduling, fleet management, and infrastructure rather than pure modeling.

What core technologies are required for this role?

Required technologies include Python, PyTorch, Hugging Face, vLLM, SGLang, TensorRT-LLM, Linux, Docker, Kubernetes, Redis, and S3-compatible storage.

What is the mission of Luma AI?

Luma AI's mission is to build unified general intelligence that can generate, understand, and operate in the physical world, emphasizing multimodality and vision.

How should I submit my application?

Applicants should apply on the official Luma AI careers page.

Ready to apply?

The application is on the official Luma AI careers page.

Apply on company site