CommonAI CIC is seeking a GPU Kernel and Performance Engineer to accelerate research models on NVIDIA GPUs, focusing on sparse and structured components. You will adapt open-source kernels, profile workloads, and optimize memory usage on cutting-edge hardware.
The role involves building tooling to measure data movement per token, validating cost models against hardware measurements, and collaborating with training infra engineers to improve throughput across diverse AI accelerators.
#J-18808-Ljbffr
GPU Kernel & Performance Engineer: Accelerate LLMs on GPUs in Cambridge employer: CommonAI C.I.C.
CommonAI is an exceptional employer, offering a collaborative and supportive engineering environment where your contributions directly influence groundbreaking AI and cloud infrastructure products. Located in Cambridge, just minutes from the railway station, we provide competitive salaries, stock options, and professional development opportunities, all while fostering a culture that values innovation and teamwork.
