Latest AI JobsBrowse Jobs
Luma AI logo

Research Scientist / Engineer – Performance Optimization

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

Luma AI is hiring a full-time Research Scientist / Engineer – Performance Optimization in London, UK, to accelerate multimodal models for efficient training and deployment. This position suits an expert in GPU optimization and transformer internals who wants to work on deep hardware performance engineering.

Role Overview

The person in this role will ensure that Luma AI's multimodal models run efficiently at scale. Responsibilities span profiling, writing custom hardware-level code, and preparing models for distributed multi-node production.

Key Responsibilities

Profile and optimize GPU, CPU, and accelerator code to achieve maximum utilization and lower latency.
Write high-performance PyTorch, Triton, and CUDA code, including custom operations.
Create fused kernels and utilize tensor cores and modern hardware features across platforms.
Adapt model implementations and architectures for distributed multi-node production environments.
Construct monitoring systems, analysis utilities, and automation for performance tracking.
Research and apply advanced optimization methods specifically for transformer models.

Requirements

Expert-level proficiency in CUDA and Triton programming along with GPU optimization.
Strong capability in PyTorch, covering custom operations and kernel creation.
Experience using profiling tools such as NVIDIA Nsight, torch profiler, and custom tooling.
Comprehensive knowledge of transformer architectures and attention mechanisms.

Nice to Have

Background with compilers and exporters including torch.compile, TensorRT, ONNX, and XLA.
Experience in tuning inference workloads to improve latency and throughput.
Familiarity with Triton compiler and kernel fusion strategies.
Understanding of warp-level intrinsics and advanced CUDA optimization techniques.

Skills and Tools

Python
PyTorch
Triton
CUDA
NVIDIA Nsight
Torch Profiler
TensorRT
ONNX
XLA
Torch.compile

About the Company

Luma AI focuses on creating unified general intelligence capable of generating, understanding, and operating within the physical world. The organization emphasizes multimodality, asserting that vision represents the necessary progression beyond language-only models. Luma AI operates as an equal opportunity employer.

How to Apply

Interested candidates should visit the official Luma AI careers page to submit an application for this position.

Skills and tech stack

PyTorchTritonCUDANVIDIA Nsighttorch profilertorch.compileTensorRTONNXXLA

Questions candidates ask

Where is the Research Scientist / Engineer position located?

The role is located in London, UK.

What core technical skills are required for this role?

Candidates need expert-level CUDA and Triton programming, GPU optimization, strong PyTorch skills including kernel development, proficiency with profiling tools like NVIDIA Nsight and torch profiler, and a deep understanding of transformer architectures.

What will the first 30 days involve?

The first 30 days focus on immersion and diagnosis, which includes profiling current training and inference paths to identify major performance opportunities.

What is the mission of Luma AI?

Luma AI aims to build unified general intelligence that can generate, understand, and operate in the physical world, with a strong focus on multimodality.

Are compilers and exporters required?

Experience with compilers and exporters like torch.compile, TensorRT, ONNX, and XLA is listed as a nice-to-have qualification rather than a strict requirement.

Ready to apply?

The application is on the official Luma AI careers page.

Apply on company site