Research Scientist / Engineer – Training Infrastructure

Luma AI

. Design, implement, and optimize efficient distributed training systems for models across thousands of GPUs .

Posted 10/8/2026full-timeRemoteMid-LevelSeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in designing and optimizing distributed training systems using PyTorch, with a focus on advanced parallelization techniques and GPU cluster management. Proficient in building monitoring tools and ensuring training stability and efficiency across large-scale environments.

Highest-signal resume keywords

Distributed PyTorch TrainingGPU Cluster ManagementAdvanced Parallelization TechniquesMonitoring and Debugging ToolsCommunication Libraries (NCCL, MPI)

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Distributed Training SystemsParallelization (FSDP, Tensor Parallel, Pipeline Parallel, Expert Parallel)Training Stability OptimizationResource Utilization OptimizationLinux Systems AdministrationScriptingMulti-Node TrainingContainerizationOrchestrationCloud Infrastructure

Industry Keywords

Foundation-Model TrainingLarge-Scale Training RunsNetworkingStorage SystemsDistributed-System Optimization

Tech Stack

Tools & technologies

CloudLinuxNode.jsPyTorch

About the role

Key responsibilities & impact

Design, implement, and optimize efficient distributed training systems for models across thousands of GPUs
Research and implement advanced parallelization, including FSDP, Tensor Parallel, Pipeline Parallel, and Expert Parallel
Build monitoring, visualization, and debugging tools for large-scale training runs
Optimize training stability, convergence, and resource utilization across massive clusters
Learn the current training stack and diagnose stability and utilization issues at scale
Deliver a parallelization or stability improvement that measurably benefits a real training run
Build monitoring and tooling to keep large runs reliable and efficient

Requirements

What you’ll need

Extensive distributed PyTorch training and parallelisms in foundation-model training
Deep understanding of GPU clusters, networking, and storage systems
Familiarity with communication libraries (NCCL, MPI) and distributed-system optimization
Strong Linux systems administration and scripting (nice to have)
Experience managing training runs across 100+ GPUs (nice to have)
Experience with containerization, orchestration, and cloud infrastructure (nice to have)
Experience at the level of FSDP and multi-node training
Ability to work remotely in the EU

Benefits

Comp & perks

Equal opportunity employer
Voluntary diversity and inclusion survey participation; refusal will not affect the job application
Remote work arrangement