GPU Kernel Engineer - High-Performance ML Acceleration

Baseten; Seattle (WA); Presencial; USD 180.000 - 360.000; Jornada completa

Descripción de la vacante

Baseten is seeking a GPU Kernel Engineer to push the performance frontiers of state-of-the-art AI workloads. You will design high-performance kernels, implement CUDA-based optimizations, and collaborate with research teams to productionize advances.

In a fast-paced environment, you will optimize memory access patterns, leverage tensor cores, and apply advanced profiling tools to reduce latency across production systems used by millions of users. This role offers strong growth and impact.

Formación

• Strong understanding of GPU architecture and programming paradigms.
• Proficient in C++ and GPU performance profiling tools.
• Knowledge of CUDA C++ API, memory access patterns, and tensor core optimization.

Responsabilidades

• Design and implement high-performance GPU kernels for key ML operations.
• Write and optimize code using CUDA, PTX assembly, and architecture-specific techniques.
• Apply advanced performance optimization methods and collaborate with research teams to productionize advancements.

Conocimientos

GPU architecture

C++ programming

CUDA

Performance profiling

Parallel programming

Tensor cores

Herramientas

Nsight Systems

Nsight Compute

Torch Profiler

Cutlass

Triton

Thrust

CUB

Descripción del empleo

Baseten is seeking a GPU Kernel Engineer to push the performance frontiers of state-of-the-art AI workloads. You will design high-performance kernels, implement CUDA-based optimizations, and collaborate with research teams to productionize advances.

In a fast-paced environment, you will optimize memory access patterns, leverage tensor cores, and apply advanced profiling tools to reduce latency across production systems used by millions of users. This role offers strong growth and impact.