Senior HPC and AI Network Software Architect
Design and evolve scalable software architectures for distributed AI training and inference, focusing on communication efficiency, runtime systems, and hardware-software co-design across GPUs, DPUs, and interconnects. Develop and optimize communication libraries like NCCL, UCX, and UCC, and collaborate with AI framework and platform teams to enhance end-to-end performance at scale. Drive innovation in AI infrastructure by integrating new protocols and system-level capabilities for large-scale models.