Senior GPU Kernel Architect & Optimizer

CoreWeave; Sunnyvale (CA); On-site; USD 182,000 - 242,000; Full time

Benefits offered by this job

Medical–dental–vision insurance

401(k) with employer match

Flexible PTO

Company-paid Life Insurance

Tuition Reimbursement

Employee stock purchase program

Job summary

CoreWeave is seeking a Senior Engineer for the Benchmarking & Performance team to own CUDA kernel authoring and optimization for AI inference. You will write, profile, and tune kernels on the critical path of large‑scale model serving to maximize throughput and minimize latency.

You will collaborate with product, orchestration, and hardware teams to deliver end‑to‑end performance gains, drive reproducible benchmarks such as MLPerf, and mentor junior engineers while upholding high coding and

Qualifications

• 5+ years in HPC, GPU software, or performance‑critical systems.
• Hands‑on CUDA experience with kernel writing and optimization.
• Deep understanding of GPU architecture, tensor cores, and memory hierarchy.
• Strong C++ and Python coding skills; low‑level performance focus.
• Familiarity with model serving stacks and kernel‑dominant costs.

Responsibilities

• Author, profile, and optimize CUDA kernels on the critical path of large‑scale model serving.
• Tune kernels for throughput and latency, leveraging tensor cores and memory optimizations.
• Prototype using kernel‑authoring DSLs and maintain reproducible benchmarking workflows.
• Lead design reviews, drive architecture within the team, and mentor juniors.
• Ensure end‑to‑end performance gains across inference stacks and meet P99 SLAs.

Skills

CUDA kernels

C++

Python

GPU architecture

Performance tuning

Tools

Nsight Compute

Nsight Systems

MLPerf

Job description

CoreWeave is seeking a Senior Engineer for the Benchmarking & Performance team to own CUDA kernel authoring and optimization for AI inference. You will write, profile, and tune kernels on the critical path of large‑scale model serving to maximize throughput and minimize latency.

You will collaborate with product, orchestration, and hardware teams to deliver end‑to‑end performance gains, drive reproducible benchmarks such as MLPerf, and mentor junior engineers while upholding high coding and