CUDA Engineer

To support a growing renewable energy startup, the full-time remote CUDA Engineer will write and optimize low-level GPU code for inference workloads, focusing on designing custom CUDA kernels and enhancing performance across various bottlenecks. Key responsibilities Write and optimize custom CUDA kernels for core transformer inference operations Profile kernels to identify and eliminate bottlenecks in occupancy, memory throughput, and warp divergence Benchmark kernels against existing baselines to drive measurable throughput and latency improvements Required qualifications 4+ years of experience writing production CUDA code with a focus on performance-critical kernels Deep understanding of GPU microarchitecture, warps, occupancy, register pressure, and memory hierarchy Strong CUDA C++ skills, including streams and asynchronous execution Hands-on experience profiling to diagnose compute-bound vs. memory-bound bottlenecks Experience with kernel fusion, memory coalescing, and avoiding warp divergence