Research Scientist / Engineer – Training Infrastructure
About this role
Luma AI is hiring a full-time Research Scientist / Engineer – Training Infrastructure in Redwood City, CA, to build distributed systems for large-scale multimodal models, making this role ideal for engineers with proven experience training foundation models at scale. The professional joining this position will focus on creating reliable, scalable, and efficient infrastructure to support advanced artificial intelligence research.
Role Overview
This position involves building the distributed systems necessary to train large-scale multimodal models across thousands of GPUs. The work requires deep technical focus on PyTorch, CUDA, and distributed systems, emphasizing advanced parallelism, training stability, and hardware utilization in massive clusters.
Key Responsibilities
Requirements
Nice to Have
Skills and Tools
Salary and Benefits
The salary range is $195K to $395K with equity offers included.
About the Company
Luma AI operates with a mission to develop unified general intelligence capable of generating, understanding, and operating within the physical world, viewing vision-based multimodality as the crucial step beyond language models. The organization functions as an equal opportunity employer.
How to Apply
Interested candidates should apply directly on the official Luma AI careers page.
Skills and tech stack
Questions candidates ask
What is the location of the Research Scientist / Engineer – Training Infrastructure role?
The position is located in Redwood City, CA.
What technologies and frameworks are central to this position?
The role heavily utilizes PyTorch, CUDA, FSDP, Tensor Parallel, Pipeline Parallel, Expert Parallel, NCCL, and MPI.
What is the compensation package for this job?
The salary ranges from $195K to $395K and includes offers of equity.
What core objective does Luma AI pursue?
Luma AI aims to build unified general intelligence that can generate, understand, and operate in the physical world through multimodal intelligence.