Quick overview
Job description
About the Role
Join a technical team building efficient training and inference systems for image, video, and world-model workloads. You will improve performance across GPU kernels, distributed engines, and multi-node infrastructure, helping make demanding AI workloads faster and more efficient.
What You'll Do
Optimize system and GPU performance for training and inference across image, video, and world-model workloads.
Profile bottlenecks at the kernel, memory, system, and cluster levels using Nsight and related tools.
Implement low-level optimizations with CUDA and Triton.
Design and improve distributed training and inference engines for diffusion models.
Improve communication performance across GPUs and nodes using NCCL, RDMA, InfiniBand, RoCE, and disaggregated serving.
Build benchmarking and regression harnesses to sustain performance improvements in production.
Co-design kernels, runtimes, and models with hardware constraints in mind.
What We're Looking For
At least 2 years of experience with deep learning training or inference systems, distributed systems, or related ML systems work.
At least 1 year of hands-on ML systems or GPU performance experience in a full-time role.
Experience authoring core features in a training or inference framework, rather than only deploying or integrating one.
Directly written NVIDIA GPU kernels using CUDA, CUTLASS, Triton, or PTX.
Experience improving inference or training performance through kernels, runtimes, or distributed execution.
Strong computer science and machine learning fundamentals, with a degree in computer science or a related quantitative field.
Experience optimizing diffusion, video, image, or other multimodal workloads is relevant.
Compensation & Benefits
Annual base salary: $180,000 to $260,000 USD. Visa sponsorship is available.
Location
On-site in Menlo Park, California, United States.