Senior Software Engineer (GPU Kernel Authoring & Optimization)

Join CoreWeave, the leading AI-cloud for high-performance GPU infrastructure. As a Senior Software Engineer, you will focus on kernel authoring and optimization, writing, profiling, and tuning GPU kernels for large-scale model serving. You will lead designs, raise engineering standards, and deliver measurable improvements to latency, throughput, and reliability across our inference stack.

• Author, profile, and optimize CUDA kernels for large-scale model serving, focusing on maximizing throughput and minimizing latency.
• Lead design reviews, drive architecture within the team, and mentor junior engineers to elevate coding/testing standards.
• Implement and maintain benchmarking workflows for end-to-end MLPerf Inference (and Training) runs, ensuring reproducibility and documentation.
• Strong coding in C++ and Python; comfortable reading and writing low-level, performance-sensitive code
• Deep understanding of GPU architecture and performance: tensor cores, warp/occupancy tuning, the memory hierarchy and bandwidth, NVLink/PCIe, and profiling with Nsight Compute/Systems
• Familiarity with model-serving stacks (vLLM, TensorRT-LLM, llm-d, SGLang) and the kernels that dominate their inference cost
• Strong communicator comfortable collaborating with cross-functional teams and external partners
• Hands-on CUDA experience is required—you have written and optimized custom kernels and are fluent with the CUDA programming and memory model
• 5+ years of experience building high-performance computing, GPU/accelerator software, or performance-critical systems
• Contributions to OSS projects such as vLLM, SGLang, PyTorch, Triton, or CUTLASS
• NCCL and collective-communication performance
• HIP / ROCm and AMD GPU experience
• Experience with SUNK (Slurm on Kubernetes) / Slurm for scheduling large GPU jobs
• CuTe DSL for Python-based kernel authoring on NVIDIA GPUs
• Experience with alternative accelerators such as Google TPUs and Meta's MTIA
• Experience running MLPerf submissions or similar large-scale audited benchmarks
• Triton or Mojo for authoring custom GPU kernels — highly desired
• Familiarity with kernel-authoring DSLs and nano-compilers such as KNYFE and its Block DSL
• JAX and its Pallas kernel language for authoring kernels on GPU/TPU
• Experience with Kubernetes at production scale
• We believe in investing in our people, and value candidates who can bring their own diversified experiences to our teams – even if you aren't a 100% skill or experience match