CUDA Kernel Engineer (Remote US)

📋 Description

• • Design, implement, and optimize custom CUDA kernels for NVIDIA GPUs to maximize occupancy, memory throughput, and warp efficiency for high-throughput AI systems serving Fortune 500 customers.
• • Profile GPU workloads using Nsight Compute, Nsight Systems, nvprof, and CUDA-MEMCHECK to analyze and eliminate performance bottlenecks such as warp divergence, uncoalesced memory access, register pressure, and PCIe transfer overhead.
• • Collaborate with AI systems, model acceleration, and backend distributed systems teams to improve GPU memory pipelines and contribute to kernel libraries and performance-engineering best practices.
• • Work on the GPU performance layer of a fast-growing AI startup founded by MIT CSAIL researchers, directly influencing the scalability and efficiency of mission-critical AI systems at scale.