Senior System Software Architect, HPC and AI Networking

NVIDIA, the world leader in accelerated computing, is seeking a Senior System Software Architect specializing in HPC and AI Networking. This role sits at the intersection of high-performance computing and artificial intelligence, focusing on building next-generation software and hardware systems for the most demanding AI workloads.

The position involves designing and implementing scalable software systems that optimize distributed AI training and inference, with a particular focus on throughput, latency, and memory efficiency. You'll work with cutting-edge communication libraries like NCCL, UCX, and UCC, collaborating with major AI framework teams including TensorFlow, PyTorch, and JAX to enhance performance and reliability.

As a Senior System Software Architect, you'll be instrumental in co-designing hardware features for GPUs, DPUs, and interconnects, directly contributing to the evolution of runtime systems and AI-specific protocol layers. The role requires deep expertise in DNNs, scaling, and parallelism, combined with strong programming skills and system architecture knowledge.

NVIDIA offers a dynamic, innovative environment where you'll work with world-class researchers and engineers. The company's commitment to fostering diversity and inclusion, combined with its position at the forefront of AI and accelerated computing, makes this an exceptional opportunity for someone passionate about shaping the future of technology.

The role is based in Beijing, China, with a hybrid work arrangement, offering the flexibility of modern work practices while maintaining collaborative opportunities with global teams. This position represents a unique chance to impact the future of AI infrastructure at one of technology's most respected and innovative companies.