Research Scientist / Engineer – Performance Optimization

Luma; New York (NY); On-site; USD 195,000 - 395,000; Full time

Job summary

Luma seeks an expert to accelerate its multimodal models by profiling GPU/CPU/accelerator code, writing high-performance kernels, and deploying at scale without compromising quality.

You will implement fused kernels, leverage tensor cores, and optimize transformer architectures for distributed multi-node production, focusing on latency, throughput, and tooling.

Qualifications

• Expert level Triton/CUDA programming for performance.
• Experience with transformer models and attention.
• Strong PyTorch kernel development and custom ops.
• Proficiency with profiling tools and performance analysis.

Responsibilities

• Profile and optimize GPU/CPU/accelerator code for max utilization and low latency.
• Develop fused kernels and leverage tensor cores across platforms.
• Write high-performance PyTorch, Triton, and CUDA code with custom ops.
• Optimize model architectures for distributed multi-node deployment.
• Build performance monitoring, automation, and tooling.
• Research and implement cutting-edge optimization techniques for transformer models.

Skills

Triton/CUDA programming

PyTorch kernel development

Transformer architectures

Profiling & optimization

Tools

Nsight

Torch Profiler

TensorRT

Custom tooling

Job description

You'll make Luma's multimodal models fast — profiling and optimizing GPU, CPU, and accelerator code so they train efficiently and deploy at scale without sacrificing quality. You'll write the kernels and operations that get the most out of the hardware. This is deep performance work: fused kernels, tensor cores, Triton and CUDA, distributed multi-node deployment. It fits someone with expert GPU-optimization skills and a deep understanding of transformer internals. If you're not at home in CUDA, Triton, and profilers, this is the wrong depth.

What You'll Own

• Profile and optimize GPU/CPU/accelerator code for maximum utilization and minimal latency.
• Write high-performance PyTorch, Triton, and CUDA, dropping to custom operations when needed.
• Develop fused kernels and leverage tensor cores and modern hardware features across platforms.
• Optimize model architectures and implementations for distributed multi-node production deployment.
• Build performance monitoring and analysis tools and automation.
• Research and implement cutting-edge optimization techniques for transformer models.

First 90 Days

One way the first 90 could unfold.

• Days 1–30 — Immerse & Diagnose: Profile the current training and inference paths and find the biggest performance wins.
• Days 30–60 — Ship & Validate: Land a kernel or architecture optimization that measurably improves utilization or latency.
• Days 60–90 — Scale & Systemize: Build the monitoring and automation that keeps performance gains from regressing.

What You Bring

• Expert-level Triton/CUDA programming and GPU optimization.
• Strong PyTorch skills, including kernel development and custom operations.
• Proficiency with profiling tools (NVIDIA Nsight, torch profiler, custom tooling).
• Deep understanding of transformer architectures and attention mechanisms.

Nice to Have

• Experience with compilers and exporters (torch.compile, TensorRT, ONNX, XLA).
• Experience optimizing inference workloads for latency and throughput.
• Triton compiler and kernel fusion techniques.
• Knowledge of warp-level intrinsics and advanced CUDA optimization.

About Luma: Luma's mission is to build unified general intelligence that can generate, understand, and operate in the physical world. We believe multimodality is critical for intelligence — the next step beyond language models comes from vision. Luma is an equal opportunity employer.

Compensation Range: $195K - $395K